Home

HKL Manual

1. a ee ai 3 3 k i is w i A eMinor amp Z O0twinow i Prof fit R Toggles turns on off the profile fitting radius You must have already indexed the image for this to be of any use because the preds shown in the display are those which are stronger A DENZO SESSION than the WEAK LEVEL and the WEAK LEVEL is determined by running Denzo The profile fitting radius is changed by entering it in the Denzo Window See PROFILE FITTING RADI US in the Denzo keyword documentation Toggles between black amp white and color Mono is often combined with reverse imaging to get black spots on a white background Launches the zoom window Any time you click the middle mouse button in either the regular or the zoom window that point becomes the new center of the zoom display Toggles between updating and removing predictions Once update predictions is turned on new predictions from each Denzo cycle will be updated automatically in the display This tells the program to write out a file which contains the current image in the xwd format You may use a local utility to convert this file into a postscript file Toggles this function on and off To use click on aD test then use the right mouse button to bracket a region of the diffraction pattern A black and red window will appear with a histogram of pixel intensities This plot but not the window can be expanded or contracted by clicking on the arrow buttons This func
2. INITIAL B FACTOR Table of starting B factors one per batch beginning on the record after the title After running the program once the output table of B factors can be cut out and pasted in here for example when you have set the FIX B flag and are not refining B values anymore 128 format default example INITIAL SCALE THE SCALEPACK KEYWORD DESCRIPTIONS INITIAL B FACTOR batch no b value batch no b value value of the default B factor which is defaulted to O INITIAL B FACTOR 1 0 0 2 0 1 3 0 1 4 0 1 5 0 2 etc Table of starting scale factors one per batch beginning on the record after the title After one run of the program the output table of scale factors can be cut out and pasted in here This table is not required If it is not included in the control input then the DEFAULT SCALE is used this is 1 unless otherwise specified If the initial scale factor is set to zero that frame is ignored in the scaling and refinement format default example INPUT format default example MERGE format default MOSAICITY format default example NO ANOMALOUS format batch no scale value batch no scale value value of the default scale factor which is defaulted to 1 INITIAL SCALE FACTOR 1 1 0 2 0 9 3 0 0 4 0 85 5 1 1 etc Redirection Tells the program to read the file which follows the keyword Same as in Denzo filename no redirection reject Flag that tells the pr
3. mar345 345mm 150 mar345 345mm 150 mar345 345mm 150 marpck mar345 345mm 150 marpck ccd unsupported m165 ccd unsupported m165 MAR CCD detector 130mm ccd unsupported mar ccd unsupported mar RAXIS IIc Rigaku scanner 105 micron INST_RX raxis raxis raster scan RAXIS IIc Rigaku scanner 210 micron INST _RXS raxis 210 raxis 210 raster scan RAXIS IIlc Rigaku scanner 105 micron raxis2n raxis2n scan Japanese version RAXIS Ilc Rigaku scanner 210 micron raxis2n 210 raxis2n 210 scan Japanese version RAXIS IV Rigaku prototype 100 micr raxis4 old raxis4 old raster scan RAXIS IV Rigaku scanner 50 micron raxis4 50 raxis4 50 raster scan RAXIS IV Rigaku scanner 100 micron raxis4 100 raxis4 100 raster scan RAXIS IV Rigaku scanner 200 micron raxis4 200 raxis4 200 raster scan STOE scanner stoe stoe CCD detector EEV chip 576 pixels ccd eev 576 ccd eev 576 CCD detector EEV chip 576 pixels double image mode ccd eev 576 double ccd eev 576 double CCD detector EEV chip 620 pixels ccd eev 620 ccd eev 620 CCD detector EEV chip 621 pixels ccd eev 621 ccd eev 621 CCD detector EEV chip 625 pixels ccd eev 625 ccd eev 625 CCD detector EEV chip 625 pixels binned CCD detector EEV chip 1152 pixels Photon Factory Amemiya drum scanner ccd eev 625 binned ccd eev pf beam line 14a ccd eev 625 binned ccd eev pf beam line 14a Photon factory fuji BA 100 scanner NL fuji nl fuji nl format Photon factory fuji BA 100 scanner NL wei
4. A wavelength 33 Oo standard variation of intensity 14 37 87 89 104 105 106 136 F 2 149 output file with On op 112 SI GMA CUTOFF Scalepack keyword 137
5. Imax 55205 4 4 44 aa 4 tag TG mag l 343 att thttagy T 4 k 19 1258 a ae er oh a arahe ee ta ks l EE 444 a a a a EN a 4 y k n A y gt eee ets 4 i T Processi 2 3 a i aN 4 4 i THE HKL MANUAL A Description of the Programs Denzo XDisplay F Scalepack An Oscillation Data Processing Suite for Macromolecular Crystallography by Daniel Gewirth written with the cooperation of the program authors Zbyszek Otwinowski and Wladek Minor revised and updated by Wladyslaw Majewski Edition 6 September 2003 Table of Contents Table of Contents Preface Introduction Data Collection Process Experimental Procedure Detector Summary The Denzo Manual Denzo for Dummys A Denzo Session Displaying the Image Using XdisplayF Visualisation of the diffraction space Starting Denzo Getting that First Orientation Indexing Indexing Refinement Using the Fit Command Mosaicity and Spot Shape How it works Integration of the Diffraction Maxima Ice Rings Refinement Statistics Finishing Up Understanding the DENZO Log File Understanding the DENZO x file Input Formats and Conventions Denzo Orientation Conventions Detector Conventions Camera Conventions Determining the values of FILM ROTATION and Y SCALE Alphabetical Denzo Keyword Descriptions Denzo Appendixes Keywords by Category Default Parameters Recommended Hardware Manual Indexing of Diffraction Images Film
6. The HKL Manual Reindex four P3 data sets each with inconsistent index and combine 110 This scenario shows how to reindex four P3 data sets and combine them This is slightly more complicated than the scenarios above Again the procedure is as follows choose one of your 4 data sets as the reference set This reference is only for the purposes of indexing not for finding scale factors Then what you do is to take each of the other 3 data sets one by one and scale them with the reference data set Each time you will have to try a different HKL MATRI X for the second data set until you find the one which is consistent with the reference data set Make a note of this HKL MATRI X and move on to the next data set When you have figured out the proper HKL MATRI X for each of the three other data sets then you can run this command file to reindex them and scale them all together This command file assumes that you were unlucky enough to have all four data sets in four inconsistent indices NUMBER OF ZONES 10 number of resolution shells for statistics ESTIMATED ERROR 0 04 0 04 0 04 0 04 ESTIMATED ERROR for each resolution shell 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 4 multiplicative factor applied to input REJECTION PROBABILITY 1 0e 4 don t reject until you know all data sets are in a consistent index WRI TE REJECTION FILE 0 9 reject RESOLUTION 40 3 3 SPACE GROUP P31 REFERENCE BATCH 201 ADD PARTIALS 1 to
7. Yes you are correct The printout definition of axes are arbitrary and are different from display direction definition The question I have concerns the choosing of box size and background size I have just completed processing data taken from a CCD detector where the box size was 75 75 the spot elliptical was 3 3 and the spot background was 35 35 Data from 2 crystals was processed using these parameters and yielded R merge values that were good However I have recently read in documentation of Denzo that the background area should be twice that of the spot area Why does this ratio of background area to spot area have to be so high Is it possible to get by with a lower ratio of background to spot area Should I go back and reprocess my data with a larger background area even though it has processed well For imaging plates does the background area still have to be twice that of the spot area Your background area was about 5 times the spot area You had a large safety margin You may get by with lower ratio at the cost of data quality XdisplayF You should click Integr box in zoom window Now if you will type in Denzo command SPOT elliptical 0 4 0 4 20 or command BACKGROUND elliptical 0 5 0 5 20 you will see integration area on the screen The spot region should include whole Bragg peak in the case of kalphal kalpha2 split box should include both peaks The spot shape is defined by statement such as SPOT elliptical 0 6 0 5 2 0 G
8. format default OMIT modifiers formats default examples OSCILLATION modifiers formats default 62 Specifies the size of the data file Equal to the number of vertical fast scanning direction lines The default value is now hard wired into the FORMAT modifier In other words once you specify the detector the NUMBER OF RECORDS value is input from a look up table NUMBER OF RECORDS integer comes along with whatever FORMAT is specified Defines the area outside a specified circle or ellipse but within the box to be omitted from any use By default there is no omitted area Defines the omit boundary to be circular with a radius value in mm The center of the omit boundary and that of the box coincide radius or rad elliptical or ell Defines the omit boundary to be ellipsoidal with characteristic major and minor semi axis lengths in mm along with the clockwise angle between a horizontal line and the major semi axis Refer to the diagram under the keyword SPOT OMIT radius value OMIT elliptical valuemajor valueminor valueangle no area omitted OMIT rad 0 35 OMIT ell 0 70 0 60 0 0 This keyword tells the program that the oscillation method is being used start Starting oscillation angle end Ending oscillation angle step Angle between Ist oscillation start and 2nd oscillation start range The actual oscillation angle OSCI LLATI ON modifier2 value2_ degrees modifierl v
9. is zero Make sure you adjust the BACKGROUND to match the size of the spot In our example above a reasonable value for the background is 0 1 mm larger than the spot in all dimensions 1 e BACKGROUND 0 7 0 6 0 0 GO In addition to adjusting the spot shape and size it is also a good idea to check and see if the PROFILE FITTING RADIUS is set properly This is easy to do with the ImageWindow In the ImageWindow simply click on the and when you move the mouse you will see a white circle which corresponds to the PROFILE FITTING RADIUS set in Denzo The pred display changes now too Instead of seeing all of the predicted reflections you see only the preds that correspond to those reflections that are stronger than the weak level Only this subset is shown because these are the only reflections used in the calculation of the average spot profile within the radius Your radius must be large enough to include enough strong reflections for proper profile fitting and not too large to average out significant profile variations As a rule of thumb you want the radius to include between 10 and 50 non weak reflections Useful increments of the profile fitting radius are 5 of the detector size REFINEMENT How it works Integration of the Diffraction Maxima Profile Fitting To calculate the diffraction intensity the detector background has to be estimated and then subtracted from the reflection profile The standard method used to estimate the
10. 36 spot radius 60 SPOT Denzo keyword 29 70 143 spot profile 9 spot separation 8 spot shape 35 Spot Shape 37 spots ellipsoidal 36 spread 8 spurious peaks 29 SSRL 155 start refinement 157 START REFINEMENT Denzo keyword 70 starting point 19 statistic error measures problems with no good statistical criteria 10 statistic error measures x weighted 10 Rmerge unweighted 10 Statistical error 95 still images 29 STOP Denzo keyword 71 structure solving 10 successfully indexed 78 sulfur 8 summary 34 Sun 77 superlattice 29 31 SWAP BYTES Denzo keyword 71 Swap space 17 Swap space 146 synchrotrons 9 28 42 Brookhaven 81 CHESS 81 115 Daresbury 155 Hamburg 156 Photon Factory 53 56 81 T Text strings 46 thin frames 137 Thin or few frames 116 TITLE Denzo keyword 71 TOP MARGI N Denzo keyword 71 total error 95 triangle 59 TRUNCATE 138 U unit cell 8 cell reduction 24 UNIT CELL Denzo keyword 53 71 UNIT CELL Scalepack keyword 139 Unix 77 unsufficient space Denzo message 154 Update Pred XdisplayF button 19 USE BEAM Denzo keyword 42 62 71 USE FIDUCI ALS Denzo keyword 72 USE NO BEAM Denzo keyword 42 INDEX USE NO FIDUCI ALS Denzo keyword 72 USE PARTIALS NO POSITION Denzo keyword 72 USE PARTIALS POSITION Denzo keyword 72 V varying mosaic spread 153 vertical axis 78 VERTICAL AXIS Denzo keyword 54 72 78 VERTICAL AXIS Scalepack keyword 13
11. For any assumed starting value of the beam position the origin of the diffraction space during indexing will be shifted to the nearest grid point of the best primitive lattice An initial error in the direct beam position by 0 48 times the distance between reflections will lead to correct indexing while an error of 0 52 times the same distance will cause a misindexing of the diffraction pattern by one index Misindexing by one is never corrected by subsequent refinement of the crystal and detector parameters Misindexing often produces poor agreement between the predicted and the observed positions of the reflections but for some crystal orientations the agreement between the predicted and the observed positions can be equally good for both correctly indexed and misindexed cases This property of the diffraction geometry creates a potential trap for the unwary crystallographer Sometimes the crystal asymmetric unit may have molecules related by an approximate translation by a fraction typically one half of a unit cell edge or diagonal The resulting diffraction pattern will have odd index reflections much weaker than even index reflections Autoindexing may find one of the two possible solutions the choice depending on whether odd reflections are weak enough to be assumed within experimental error systematically absent or not This depends on what is the fraction of odd reflections in the peak search If there are only a few odd reflections th
12. What follows is a description of how you would continue from the lattice type given by Denzo to determine your space group This whole analysis of course only applies to enantiomorphic compounds like proteins It does not apply to small molecules necessarily which may crystallize in centrosymmetric space groups If you expect a centrosymmetric space group you use any space group which is a subgroup of the Laue category to which your crystal belongs You also need enough data for this analysis to work so that you can see systematic absences To determine your space group follow these steps L 2 Determine the lattice type in Denzo Scale by the primary space group in Scalepack The primary space groups are the first space groups in each Bravais lattice type in the table which follows In the absence of lattice pseudosymmetries e g monoclinic with B 90 the primary space group will not incorrectly relate symmetry related reflections Note the y7 statistics Now try a higher symmetry space group next down the list and repeat the scaling keeping everything else the same If the y is about the same then you know that this is OK and you can continue If the y are much worse then you know that this the wrong space group and the previous choice was your space group The exception is primitive hexagonal where you should try P6 after failing P3 21 and P3 12 Examine the bottom of the log file for the list of candidate systematic
13. batten roty 1 Aso 180 FIT batch mosaicity 1 to 180 FORMAT denzo_ip HKL MATRIX O 1 O reindexing matrix h k 0 Oo 1 k 1 LO 30 I h SECTOR 1 to 180 FILE 1 frames myxtal x OUTPUT FILE myxtal sca OUTPUT FILE contains reindexed reflections 103 The HKL Manual Convert a b cto c a bin orthorhombic NUMBER OF ZONES 10 number of resolution shells for statistics ESTIMATED ERROR 0 04 0 04 0 04 0 04 ESTIMATED ERROR for each resolution shell 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 4 Multiplicative factor applied to input o REJECTION PROBABILITY 0 0001 WRITE REJECTION FILE 0 9 reject RESOLUTION 40 3 3 SPACE GROUP P21212 REFERENCE BATCH 1 ADD PARTIALS 1 to 180 POSTREFINE 10 FIT erystal a T vor 160 Rll emystad b Lto 180 FT crystal c L to 180 PiE batch rocs J te 1o00 FIT pater rocy L C6 380 FLT batei mosaicity 1 26 gt 4 380 FORMAT denzo_ip HKL MATRIX O 0 1 reindexing matrix h 1 Te 0 0 kIT h ce ee LIT k SECTOR 1 to 180 FILE 1 frames Denzo myxtal x OUTPUT FILE myxtal sca OUTPUT FILE contains reindexed reflections 104 Reindex P23 switch a b NUMBER OF ZONES 10 ESTIMATED ERROR 0 04 0 04 0 04 0 04 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 4 REJECTION PROBABILITY 1 0e 4 WRI TE REJECTION FILE 0 9 reject RESOLUTION 40 3 3 SPACE GROUP P23 REFERENCE BATCH 1001 ADD PARTIALS 1 to 180 1001 to 1180 POSTREFINE 10
14. ccd adsc unsupported q315 ccd adsc unsupported q315 binned ccd adsc mar_corr mar Ibl mar compressed mar 18cm mar 30cm mar345 180mm 100 mar345 180mm 100 marpck mar345 180mm 150 mar345 180mm 150 marpck mar345 240mm 100 mar345 240mm 100 marpck mar345 240mm 150 mar345 240mm 150 marpck mar345 300mm 100 mar345 300mm 100 marpck mar345 300mm 150 mar345 300mm 150 marpck Denzo Format Modifier ccd adsc quantum4 binned ccd adsc quantum4 binned dz ccd adsc unsupported q4 ccd adsc binned unsupported q4 ccd adsc unsupported q210 ccd adsc unsupported q210 binned ccd adsc unsupported q315 ccd adsc unsupported q315 binned ccd adsc mar_corr mar Ibl mar compressed mar 18cm mar 30cm mar345 180mm 100 mar345 180mm 100 marpck mar345 180mm 150 mar345 180mm 150 marpck mar345 240mm 100 mar345 240mm 100 marpck mar345 240mm 150 mar345 240mm 150 marpck mar345 300mm 100 mar345 300mm 100 marpck mar345 300mm 150 mar345 300mm 150 marpck ABOUT THE SCALEPACK MANUAL Detector Description Modifier XdisplayF Modifier XdisplayF Denzo Format Modifier Ver 1 5 and lower Ver 1 96 and higher MAR 345mm scanner 345mm scan 100 mar345 345mm 100 mar345 345mm 100 micron size MAR 345mm scanner 345mm scan 100 micron size compressed MAR 345mm scanner 345mm scan 150 micron size MAR 345mm scanner 345mm scan 150 micron size compressed MAR CCD detector 165 mm mar345 345mm 100 marpck mar345 345mm 100 marpck
15. statistics 64 zones 65 Profile Fit R XdisplayF button 18 profile fitting 14 149 profile fitting radius 18 Profile fitting 148 profile fitting radius 14 66 149 PROFILE TEST Scalepack keyword 133 PROFILES FITTED Scalepack keyword 133 163 The HKL Manual PROFILES SUMMED Scalepack keyword 134 prompt 22 pseudo symmetry 24 pseudo zone 79 Q quadrilateral 58 153 QUI T Denzo keyword 66 71 R RADIAL OFFSET Denzo keyword 33 51 66 radiation damage 96 radius of convergence 32 RAM 17 RAW DATA FILE Denzo keyword 67 R axis gt detectors 37 55 reciprocal space 22 29 31 126 recommended hardware 77 82 RECORD LENGTH Denzo keyword 67 RECSQ Scalepack keyword 134 redundancy 89 reference batch 88 REFERENCE BATCH Scalepack keyword 95 97 98 100 103 134 REFINE NO PARTI ALI TY Denzo keyword 67 REFINE PARTI ALI TY Denzo keyword 67 refine dat command file 13 refine dat command file 4l refinement 9 14 refinement Q amp A 156 strategy 6 refineone dat 33 reflection conditions 117 reindexing in Scalepack 102 reject 68 REJECT Denzo keyword 40 68 REJECT HKL Scalepack keyword 134 REJECT OUTLIERS Scalepack keyword 135 reject dat 135 rejection criteria 68 90 REJECTION PROBABILITY Scalepack keyword 135 rejects 88 remeasured 44 reprocess 35 RES Denzo keyword 68 resolution displayed in XdisplayF 21 resolution cutoff 144 resolution limit 8 90 RESOLUTI ON LIMITS Denzo keywo
16. the BOX of Denzo plus the circles which define the spot and the guard regions both in white Note that you have to have indexed the frame in Denzo for this display to be useful since the integration box and the predicted reflections are determined in Denzo not XdisplayF Even though you entered A DENZO SESSION the spot guard and box sizes in mm in Denzo their actual sizes must be an integral number of pixels so you are not going to see perfect circles Also the spot must have numerical parity with the box i e the background will always be symmetrical around the spot so if the box has an odd number of pixels in a direction then the spot will as well even if it is supposed to be circular The next time you hit the the box is removed and the spot outline is colored red This makes it easy to see things when spots are close together When you hit the Int box once more the original display is restored When this button is clicked a second black circle is displayed which is centered on the actual observed reflection The green or yellow circles representing whole or partial reflections are centered on the predicted positions of the reflections If the indexing and refinement have gone well the two circles should be very close to concentric Notice that most of the time when they are not they differ by only a little bit When they differ by a lot you will see a little line drawn which connects the centers of the obser
17. 134 136 154 B RESTRAIN Scalepack keyword 122 DEFAULT B FACTOR Scalepack keyword 122 FIT B Scalepack keyword 126 INITIAL B FACTOR Scalepack keyword 128 background 6 9 21 22 31 35 37 39 40 63 72 158 BACKGROUND CORRECT ON Scalepack keyword 90 122 BACKGROUND Denzo keyword 25 36 42 51 background overflow 44 background overflow 156 background size 145 background subtraction and fluctuation 137 BOX Denzo kayword 52 high intrinsic background 38 non random signal in background 135 REJECT fraction Denzo command 68 BAD VALUE Denzo keyword 52 batch data set in Scalepack 124 batch processing 15 16 41 81 156 batchDenzo com command file 41 film pack parameters 75 keywords required for 76 Scalepack 88 94 159 The HKL Manual SECTOR Denzo keyword 69 batchDenzo com command file 4l Bayesian reasoning 10 89 rejection of outliers 135 beam 9 angular spread 9 29 36 53 beam search routine 81 beam stop shadow 58 CROSSFIRE Denzo keyword 53 direct beam 23 28 30 53 81 efficient use 7 findbeam dat search routine 42 focus 9 53 intensity fluctuations 9 30 LAMBDA Denzo keyword 72 NO BEAM Denzo keyword 62 71 synchrotrons 61 thinned beam stop 28 81 USE BEAM Denzo keyword 42 62 71 USE NO BEAM Denzo keyword 42 WAVELENGTH Denzo keyword 25 72 wavelength modifier to FI T Scalepack keyword 125 x beam amp y beam parameters 78 X BEAMand Y BEAM positions 2
18. 207 P432 208 P4232 2n 0 0 212 P4332 4n 0 0 213 P4132 4n 0 0 Centered Cubic 1213 197 123 i 199 1213 s 14132 211 1432 214 14132 4n 0 0 F Centered Cubic F23 196 F23 F4132 209 F432 210 F4132 2n 0 0 Primitive Rhombohedral R3 146 R3 R32 155 R32 Primitive Hexagonal P31 143 P3 144 P31 0 0 3n 145 P32 0 0 3n P3112 149 P312 151 P3112 0 0 3n 153 P3212 0 0 3n P3121 150 P321 152 P3121 0 0 3n 154 P3221 0 0 3n P61 168 P6 169 P61 0 0 6n 170 P65 0 0 6n 171 P62 0 0 3n 172 P64 0 0 3n 173 P63 0 0 2n P6122 177 P622 178 P6122 0 0 6n 179 P6522 0 0 6n 180 P6222 0 0 3n 181 P6422 0 0 3n 182 P6322 0 0 2n 118 Bravais Lattice Primitive Tetragonal Centered Tetragonal Primitive Orthorhombic C Centered Orthorhombic Centered Orthorhombic F Centered Orthorhombic Primitive Monoclinic C Centered Monoclinic Primitive Triclinic Primary assigned Space Groups P41 P41212 141 14122 P212121 C2221 1212121 F222 P21 C2 P1 Candidates P4 P41 P42 P43 P422 P4212 P4122 P4322 P4222 P42212 P41212 P43212 14 141 1422 14122 P222 P2221 P21212 P212121 C2221 C222 1222 1212121 F222 P2 P21 C2 P1 SCENARIOS Reflection Conditions along screw axes 0 0 4n 0 0 2n 0 0 4n 0 2n 0 0 0 4n 0 0 4n 0 0 2n 0 0 2n 0 2n 0 0 0 4n 0 2n 0 0 0 4n 0 2n 0 0 0 4n 0 0 4n 0 0 2n 2n 0
19. Crystallographica A25 43 55 38 The profile fitting increases the precision decreases the statistical error of the measurement but it may introduce an error due to lack of accuracy of the predicted profiles Denzo applies the averaging of profiles in detector coordinates and unlike other programs that use the profile fitting method averages profiles separately for each spot This approach has two main advantages 1 first only nearby spots are chosen for averaging ones which should have the most similar profiles 2 second Denzo avoids interpolation in the profile prediction step instead it shifts the contributing profiles by vectors that make the smallest possible pixel truncation error These translation vectors precisely center the predicted profile on the reflection to be fitted and the error introduced by these shifts is smaller than that due to interpolation used in some other programs Prediction of the profile shape is never exact due to errors in the positional refinement due to averaging of different shapes due to truncation of pixel shifts or interpolation etc The resulting error of the fitted intensity was analyzed by Diamond in the case of one dimensional Gaussian profiles and an unweighted profile fitting formula The important parameters are w the root mean square rms width of the actual profile f the rms width of the predicted profile and d the displacement of the predicted profile from the actual profi
20. FIT crystal a 1 to 180 FLT Batch rots I to 130 FIT batem roty 1i to 180 FIT batch mosaicity 1 to 180 FORMAT denzo_ip HKL MATRIX O 1 0 R OO SECTOR 1 to 180 FILE 1 frames myxtal2 x Unit Matrix HKL MATRIX 1 0 O Olesen O 0 1 SECTOR 1 to 180 FILE 1001 frames myxtall x OUTPUT FILE myxtal sca SCENARIOS number of resolution shells for statistics ESTI MATED ERROR for each resolution shell multiplicative factor applied to input o commented out to spare log file in case this reindexing is incorrect reindexing matrix h k k h also works for I23 and F23 this data will be reindexed by the matrix above data set 1 to be scaled with reindexed data this data will not be reindexed output combines 2 data sets with consistent indexes 105 The HKL Manual P321 or P312 Reindex data and combine with another data set NUMBER OF ZONES 10 number of resolution shells for statistics ESTIMATED ERROR 0 04 0 04 0 04 ESTIMATED ERROR for each resolution shell 0 04 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 4 multiplicative factor applied to input o REJECTION PROBABILITY 0 0001 commented out to spare log file in case WRITE REJECTION FILE 0 9 reindexing is incorrect reject RESOLUTION 40 3 3 SPACE GROUP P321 REFERENCE BATCH 101 ADD PARTIALS 1 to 60 101 to 160 POSTREFINE 10 good idea to POSTREFI NE FIT crystal a 1 to 160 HLT erystad c Lto T60 FLT Batch
21. MAT FILE 1 native sca REFERENCE BATCH 1 REFERENCE BATCH in scaling procedure We will compare derivative to native so native is reference Derivative RESOLUTION 2 8 use highest resolution from the worse data set HKL MATRIX O O0 1 only reindex if native and derivative have different indices Oo O L 0 Q0 FILE 2 NewHg sca derivative data set OUTPUT FILE NL junk sca not interested in sca file only the log print total chiz 1000 these are set super high so that basically no outliers are printed print single chi2 1000 to log file You would use this scenario if you wanted to see how well a native and derivative data set scaled together It is different from Scenario 3 because both data sets are complete and are being read in as sca files This is a more accurate way of identifying your derivatives but it requires enough data from the derivative so that it can be internally scaled IlI The HKL Manual Scenario 7 Compare l and l detection of anomalous signal This involves a 2 step scaling procedure In the first step the data are scaled and postrefined along the lines of Scenario I with one exception noted below namely that the ANOMALOUS flag has been set This tells the program to output a sca file where the I and I reflections are separate Then in the second step a clever trick is used the sca file is read in but the ANOMALOUS flag is turned off In this case Scalepack will treat the I data and the I data as t
22. R3 in primitive setting 155 R32 R32 in hexagonal setting AS H32 R32 in primitive setting E Non standard but useful to make angles close to 90 501 Ti Non standard but useful to make angles close to 90 202 P2C P2 C axis unique 403 P21C P21 C axis unique 305 B like C2 B face centered c axis unique 503 I2 Non standard but useful to make beta angle close to 90 SPINDLE AXIS Alias for ORI ENTATI ON AXIS 2 UNIT CELL Real cell specified as a b c alpha beta gamma UNI T CELL is included in the header of Denzo_IP Denzo_york1 ARCHIVE and Scalepack files otherwise it must be included Format UNIT CELL value value value angle angle angle Default the value from the first header encountered Example UNIT CELL 50 62 100 3 90 90 90 E the postrefined value of the unit cell constants is not used nor is it output in the Scalepack output file You must get this information from the log file if you are interested in it VERTICAL AXIS Same as ORIENTATION AXIS 1 WRITE BADDIES Writes xrej files so that reflections from the reject file may be displayed by XdisplayF Not fully implemented Will terminate the program and prevent scaling and postrefinement Note that the command reject must precede the input of the x files 139 The HKL Manual WRITE REJECTION FILE Creates a file with Ak s to be rejected UNIX file name reject VMS file name REJECT DAT Reject file is created if it does not exist If the file exists it i
23. REFINEMENT Denzo Orientation Conventions The standard display and processing convention for Denzo is to show the diffraction pattern such that as you view the image from the crystal side of the detector the spindle axis is horizontal and coming in from the right hand side What follows is an explanation of how this corresponds to the detector data storage convention Detector Conventions A detector writes data to a file as a series of lines The X coordinate is defined as the line number times the pixel RASTER size and the y coordinate is defined as the pixel position within the line times the pixel size in the y direction The outer corner of the first pixel in the data file begins at x 0 y 0 Unfortunately the data file convention doesn t correspond to the laboratory frame of reference because it is not standardized among detector manufacturers Thus there is a total of 8 possible combinations of x and y conventions They are shown below x slow fast y Figure 5 The eight possible data file conventions As one might guess detector manufacturers have managed to use 7 of these 8 conventions Only 1 left SUYS The table below gives some examples of the conventions used by the major detectors supported by Denzo for a detailed list see the Detector Specifications appendix RaxisllIc upside down Kodak Kodak DIP100 Fuyji_8 DIP 3000 Mar DIP320 What really counts is not the scanni
24. SCALE FACTOR greater than say 2 0 then I know that it still isn t right I tried scaling again with ERROR SCALE FACTOR 1 5 ESTIMATED ERROR 0 10 REJECTION PROBABILITY O 1 This time the scaling statistics look like Shell Lower Upper Average Average Norm Linear Square limit Angstrom L GPLOr Stabs Chit R fac lt R fac 100 00 4 31 207375 1 177638 1 6884 6 4 242 0 088 0 136 4 31 3 42 141622 7 11636 3 3527 6 4 565 0 080 0 115 3 42 2 99 608354 7 5307 6 1921 0 4 949 0 091 0 145 iii AeA 317748 7 3013 8 1348 5 4 555 0 099 0 146 Zed eae 21800 0 2247 5 1161 2 4 118 0 109 0 146 Pees wally 162350 9 1822 3 1063 3 3 710 0 117 0 141 Bet ewe eee 1602 3 1020 6 3 642 0 128 0 160 wee Zens 104350 2 1422 9 988 6 4 094 0 157 0 206 Beko Od 8010 3 1310 4 1000 5 3 451 0 184 0 220 2 07 2 00 6584 4 1296 2 1033 9 3 515 0 220 0 268 All reflections 52826 6 4803 3 2009 7 4 142 0 093 0 131 numbers Any suggestions as to how to make this a bit better At least now all the y are similar I know that I can compensate for these by increasing ERROR SCALE FACTOR but is there a better way What exactly is the scanner doing to give such funny data Elspeth thought that there was a problem with the nonlinearity of the detector above a certain value which is why I tried using OVERLOAD VALUE in Denzo to try and get rid of rubbish data I ve fiddled around with all my data and it doesn t scale well no matter what I do It s a shame because my data is 96 complete t
25. above Unit Matrix HKL MATRIX 1 O 0 data set 1 to be scaled with reindexed data O t Os 104 ul SECTOR 1 to 60 FILE 101 frames myxtall x this data will not be reindexed OUTPUT FILE myxtals12 sca output combines 2 data sets with consistent indexes 107 The HKL Manual Reindex from H32 to R32 NUMBER OF ZONES 10 number of resolution shells for statistics ESTIMATED ERROR 0 04 0 04 ESTIMATED ERROR for each resolution shell 0 04 0 04 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 4 multiplicative factor applied to input o REJECTION PROBABILITY 1 0e 4 WRITE REJECTION FILE 0 9 reject RESOLUTION 40 3 3 SPACE GROUP R32 after reindexing this will be the space group REFERENCE BATCH 1 ADD PARTIALS 1 to 90 POSTREFINE 10 good idea to POSTREFI NE because this will affect partials FIT crystal a 1 to 90 FIT erystadl e 1 to 90 FIT pateh rots L to 90 FUT Dacth roty 1 CTo 390 FIT Batch mosaicity 1 to 90 FORMAT denzo_ip HKL MATRIX 1 0O 1 SECTOR 1 to 90 FILE 1 frames myxtal x this data will be reindexed by matrix above OUTPUT FILE myxtals12 sca output combines 2 data sets w same index 108 SCENARIOS Reindex two H3 data sets each with inconsistent index and combine This scenario shows how to reindex two H3 data sets into R3 and combine them This is slightly more complicated than the scenarios above since the second data set must also be reindexed into R3 from H3
26. absences If this was the correct space group all of these reflections should be absent and their values very small Compare this list with the listing of reflection conditions by each of the candidate space groups The set of absences seen in your data which corresponds to the absences characteristic of the listed space groups identifies your space group or pair of space groups Note that you cannot do any better than this 1 e get the handedness of screw axes without phase information If it turns out that your space group is orthorhombic and contains one or two screw axes you may need to reindex to align the screw axes with the standard definition If you have one screw axis your space group is P222 with the screw axis along c If you have two screw axes then your space group is P2 2 2 with the screw axes along a and b If the Denzo indexing is not the same as these then you should reindex using one of the two transformations described above in Scenario 5 so that you are following the standard conventions So far this is the way to index according to the conventions of the International Tables If you prefer to use a private convention you may have to work out own transformations One such transformation has been provide in the case of space groups P2 and P2 117 The HKL Manual Primary assigned peuecion Bravais Lattice Space Groups Candidates Conditions along screw axes Primitive Cubic P213 195 P23 198 P213 2n 0 0 P4132
27. based on high redundancy 4 fold or higher and high resolution 2A or better should be expected Only very few less than 0 1 outliers should be found during merging Results worse than the above indicate a problem with the test crystal or with the experimental setup Preferably the test crystal should be kept at 100K to minimize radiation damage Problems with the test crystal may mask detector problems For instance slippage of the test crystal makes it very difficult to notice a spindle motor backlash or malfunctioning of the X ray shutter Benchmarks and Calibration Many macromolecular crystallography labs have not developed stringent benchmarks of acceptable performance The value of such a test depends on how it is analyzed Frequently lack of rigor leads to the acceptance of many outliers in the test data The outliers rejection in the merging of symmetry related data is a valid statistical procedure but it should be applied with great caution The definition of an outlier is a large but sporadic fluctuation in the data for example due to a cosmic radiation hit A small number less than 0 01 of outliers is something to be expected even in a well functioning system However the practice of many labs has been to accept a much larger number of outliers even as high as 10 Many serious problems may be masked by such a liberal outlier rejection It should be emphasized that outlier rejection always improves the consistency in
28. between Especially useful for black and white displays so that the spots appear black and the background appears white Starts the peak search routine and results in the creation of a file called peaks file The number of spots 0 Peak Sear picks can be controlled with menu options which will appear on the right hand side after the initial peaks are found The picked peaks are displayed as small black circles over the image This set of peaks is the starting point for the autoindexing routine in Denzo 19 THE H KL MANUAL Right mouse button Mid mouse button Left mouse button 20 Increases the display dynamic range which has the effect of emphasizing the stronger reflections with a concomitant loss of the weaker ones It could also be called Top up to make it analogous with the Floor Down button Opposite of Lain Decreases the display dynamic range which has the effect of emphasizing the weaker pixels The dark pixels will just stay dark This is analogous to the Floor up but at the other end of the spectrum Quit and exit the display program This will not interfere with any Denzo jobs running in a separate window as long as you do not log off the computer Tells the display which direction to increment the last digit of the image file when displaying sequential images See the description above for The right mouse button causes the file number to increment and the left mouse button cause the file numb
29. binned ccd nonius ccd nonius dz ccd nonius binned ccd nonius binned dz ccd noniusb ccd noniusb dz ccd noniusb binned ccd noniusb binned dz ccd brandeis Denzo Format Modifier fuji 1 Obit film 25 film 50 film 100 film lars film is ccd bruker quad ccd bruker quad binned ccd smart6000 ccd bruker ccd smart6000 binned ccd aps1 ccd aps1 binned ccd sbc2 ccd sbc2 binned ccd nonius ccd nonius dz ccd nonius binned ccd nonius binned dz ccd noniusb ccd noniusb dz ccd noniusb binned ccd noniusb binned dz ccd brandeis ABOUT THE SCALEPACK MANUAL Detector Description Modifier XdisplayF Modifier XdisplayF Denzo Format Modifier Ver 1 5 and lower Ver 1 96 and higher CCD Brandeis 2k CCD single chip detector at X12C CCD Brandeis 2x2 CCD ESRF detector BL19 at ESRF raw data CCD ESRF detector BL19 at ESRF gt 1200x1200 pxl CCD ESRF detector BL19 at ESRF gt 1200x1200 pxl CCD ESRF detector 1024 1024 CCD ESRF detector 1152x1242 CCD 1k 55mm original Sol Gruner INST CCD detector at CHESS CCD 2k detector CHESS 2048x2048 INST_CCD2x2 image CCD 2k detector CHESS 1024x1024 binned mode MAC Science DIP 100 INST_CCD2048b INST_DIP100 MAC Science DIP 2000 12 bit A D converter MAC Science DIP 2000 16 bit A D converter MAC Science DIP 2030 12 bit A D converter MAC Science DIP 2030 16 bit A D converter MAC Science DIP 2040 16 bit A D converter MAC Science DIP 3000 cyl
30. by a Denzo keyword The FIT command is followed by the command G0 which calls one cycle of refinement One GO command calls one cycle of refinement 5 GO commands call for 5 cycles of refinement etc If a keyword parameter is fixed it will retain its current value through the refinement The program assumes that all of the keyword parameters are fixed at the beginning of every new frame The effect of these commands is cumulative and carries over through subsequent cycles of refinement For example if the command FIT CRYSTAL rotx roty rotz G0 is given then on every subsequent refinement cycle called by the GO statement the CRYSTAL rotx roty and rotz parameters will be refined Typically a refinement starts by fitting just a few parameters like CRYSTAL rotx roty rotz allowing the refinement to proceed towards convergence for a few cycles and then fitting more parameters a few at a time The keyworded parameters that can be FIT are given below All of the keyworded parameters can be refined simultaneously so care should be taken not to fit parameters which are too strongly correlated at low resolution For example the crystal to film distance and the unit cell lengths are highly correlated so one should be allowed to refine before refining the other usually the unit cell lengths are refined first The END FIT and END FIX command words mark the end of a list of fitted parameters This is especially useful for inputing more parameters after
31. close to the oscillation axis from affecting the refinement very much partiality the expected error of the partiality determination The value for the error is expressed as a decimal e g 10 would be written 0 1 systematic the weight given the positional refinement compared to the partiality refinement when refining crystal orientation parameters The higher the value the more weight is given to the positional refinement Typical values range from 1 to 20 A value of 5 seems to work almost all the time ERROR modifierl value modifier2 value overload 0 005 density detector and wavelength dependent positional 0030 y x factor 0 1 partiality Uso systematic 5 0 ERROR systematic 5 00 partiality 0 100 positional 0 050 55 THE HKL MANUAL ad FIDUCIALS format default example FILM modifiers formats defaults examples 56 Avoid modifying the error parameters other than positional and density Moreover error density should be constant for a detector site and changed only with caution Defines the position of the fiducial marks on the film relative to the beam spot before FI LM rotation andY SCALE are applied Given as a series of x y coordinates in mm These days this is only used with Photon Factory Weissenberg images FIDUCIALS xvalue_ mm yvalue_ mm xvalue_ mm yvalue_ mm none FIUDUCI ALS 90 0 90 0 90 0 90 0 This keyword tells the program about the characteristics of the data c
32. differences and want to use only the most accurate data Another time you might ignore overloads is when you collect multipass data In this case a crystal is exposed twice once for a short time the other for a longer time The longer exposure is to sufficiently darken the high resolution reflections but will result in saturated low resolution reflections Since the low resolution reflections can be obtained from the short exposures the overloaded ones can be ignored in the long exposures See Press William H Teukolsky Saul A Vettering William T Flannery Brian P Numerical Recipes in C The Art of Scientific Computing Second Edition Cambridge University Press 1992 IMPLEMENTATION Implementation Limits on variable dimensions Platforms The current standard version scalepack allows up to 2 000 000 total observations 1 000 000 unique reflections and 2000 films frames or batches The program will refine up to 3000 parameters e g scale factor B factor simultaneously This should be enough for the most uses of the program If the number of observations number of unique reflections etc exceeds the standard scalepack limits you may use a bigger version of scalepack This is usually only a problem in virus crystallography or in case of very high resolution data For this purpose three other versions of Scalepack have been made Scalepack8m limit 2x10 unique reflections Scalepackl6m 3x10 unique reflections and
33. etc etc On VMS system you can put all files in any directory You should set up logicals by including into your login com file a bunch of commands like S KODAK SZBYSZEK WLADEK DISTR VMS CORNELL EXE S FUJI SZBYSZEK WLADEK DISTR VMS FUJI EXE S RX SZBYSZEK WLADEK DISTR VMS RX EXE RXS SZBYSZEK WLADEK DISTR VMS RXS EXE MARL8 ZBYSZEK WLADEK DISTR VMS HAMBURGN EXE MAR30 SZBYSZEK WLADEK DISTR VMS HAMBURGD EXE S EMBL ZBYSZEK WLADEK DISTR VMS HAMBURG EXE DENZO SZBYSZEK WLADEK DISTR VMS DENZO EXE I am very happy with Scalepack and am trying to use it to look at data from a Xuong Hamlin detector The ar files were on a MicroVAX where the data was collected and I moved them over to an IRIS Indigo to run Scalepack I have been using the following input file but I get the following error ENDFILE END OF FILE 1 APPARENT STATE UNIT 5 NAMED LAST FORMAT 132A1 LATELY READING SEQUENTIAL FORMATTED INTERNAL IO This occurs even if I specify a non existent file after the keyword FILE Do I need to include another keyword such as file size Hamlin Archive files are binary and have to be processed with Scalepack on the same type of computer If you did reduce Hamlin data on VAX you have to run Scalepack on VAX or Alpha VMS SGI version of Scalepack does not understand VAX Hamlin binary files Sorry 147 The HKL Manual Avoid byte swapping of raw data with comman
34. file itis most likely due to an error in the site dat file Candidates include X BEAM and Y BEAM EP Notice that the session was run interactively not as a batch job This is so that you can watch the progress of the refinement in the ImageWindow Also it is to encourage you to process your images as they are collected so that you can spot problems before you have wasted lots of time If the refinement is not working you will be able to see very quickly that the predicted reflections preds do not line up with the real reflections and stop the job If the data collection has problems you can stop the experiment and fix your crystal oscillation range etc If the computer is reasonably fast the data are on a local disk and the number of reflections is average you can probably process 90 frames in less than half an hour which is not too bad If you still want a log file to examine later once you catch your breath try typing something like gt denzo tee denzo log when starting Denzo 15 THE H KL MANUAL A Denzo Session This chapter describes data processing in more detail than in Denzo for Dummys Denzo session outline Load frames onto Disk ImaqeWindow DenzoWindow Launch XDisplayF confirm find xbeam edit myexperiment dat xbeam ybeam ybeam launch Denzo peaks file site dat myexperiment dat verify that preds match spots hklpredictions amp others failure to index check refineo
35. for a given resolution range the number of pixels needed to save the spot profile and two computer memory statements REFINEMENT hklmax 99 99 99 93 pixels needed to save spot profile maximal usage of the common sfprof 283044 2 Below is a complete list of the suspect but not rejected reflections The columns contain in order index x and y coordinate of the reflection centroid the oscillation angle at which maximum diffraction occurs within the oscillation range the measured intensity of the reflection the normalized intensity of the reflection the error on I and then some flags which describe the problem with the reflection There are only two types of problems flagged negative intensity and positional error This list is useful because you can go back to the image and see what the problem is for each reflection h k a X y oscang intens scalI errI xdel errx ydel erry 27 84 5 p 60 25 31 81 0 02 797 9 0 29 129 6 0 24 1 negative 4 28 2p 96 13 91 63 0 50 1696 6 0 03 114 8 0 115 0 033 0 274 0 036 0 0 1 position 15 2 1 p 109 66 126 93 0 47 5601 3 0 03 122 1 0 019 0 031 0 129 0 031 0 3 1 position 3 Below is the mosaicity histogram for the reflections of the frame The reflections are sorted into 20 zones which range from minus the input mosaicity 2 to plus the input mosaicity 2 Each zone represents the shortest angular distance of the center of the reflection from the surface of the Ewald sphere at the end of the oscillat
36. formula Distance Denzo distance R Axis cos 29 X BEAM and Y BEAM are the position of the direct beam on the IP during exposure not at 20 0 For large values of 20 the Y BEAM may be off the detector The correct value may be negative or more than 200 in such cases CRYSTAL rotz and CASSETTE rotz are strongly correlated and cannot be refined simultaneously Alias for the keyword UNIT CELL This value reflects how fast the detector moves during exposure for Weissenberg images Given as mm per degree This is the inverse of the Photon Factory parameter degree mm and is only useful for Weissenberg images COUPLING value_ mm degree 0 0 COUPLING 0 057 Crossfire is a measure of the X ray beam divergence and focusing as it leaves the collimator and illuminates the crystal Crossfire being a symmetric tensor has x y and xy components It affects the prediction of partial reflections and their positions not their angular width It is expressed as angular divergence of the beam The default value is zero crossfire i e a perfectly parallel beam or a beam focused on the detector X angular spread of the beam in the x direction 4 angular spread of the beam in the y direction xy correlated xy component of the beam spread This tends to be zero within error CROSSFIRE x valuel_ arbitrary_units CROSSFIRE y value2_ arbitrary_units CROSSFIRE xy value3_ arbitrary_units 000 CROSSFIRE x 0 023 value taken from previous refinement CR
37. gt 10 you may want to increase the profile fitting radius Let s say you need to increase it slightly PROFILE FITTING RADIUS 30 GO Hopefully you get the idea of how to adjust your parameters Step 8 see Mosaicity and Spot Shape and Refinement Statistics in Refinement Using the Fit Command see If things get worse in Refinement Using the Fit Command 14 DenzoWindow ImageWindow refine dat begins sequential refinement of frames specified in the sector n to m section of myexperiment dat Depending on how fast your computer is and how fast you can read examine the screen output for the correct histogram shape meaning you chose the mosaicity correctly and for good y values and convergence If your site dat file was set up correctly things should work Watch refinement proceed Little circles should line up on DENZO FOR DUMMYS reflections Cool If not gotta read more manual Step 9 Denzo Window After last frame type STOP This ends the program ImageWindow Hit close button to kill image You re done Result is a bunch of xtal x files Step 10 Analysis Scroll backward through the DenzoWindow to examine the ee OI tie Dene histogram and the final y values If they are normal you can Log file in Refinement Using proceed to Scalepack If not see the discussion in the following ihe Fit Command chapter If this scenario didn t work and you entered the correct values inthe myexperiment dat
38. in the rot values This is because the shifts reported are the A not the changes in rotx roty and rotz The other reason for fitting A rotations the way they are defined in Denzo is to make them have a more intuitive correlation with the other parameters Otherwise changes in crystal rotz would not correlate with cassette rotz This is of more importance in Scalepack where only rotx and roty and not rotz are refined E About fitting unit cell parameters In both Denzo and Scalepack the unit cell is fitted in FIT B format default example FIX B format default Example FIXED WINDOW reciprocal space not real space This means that for a non orthogonal space group refining the value of a may end up changing the values of b and c even though b and c remain the same The same is true for the angles fitting a may end up changing B and y even though B and y remain the same So what you may notice sometime if you are not careful is that when you ask the program to fit crystal a but not b c and the angles then a will not be a constant a will be constant but when converted back to a when the other unit cell parameters have not been changes a will not The moral when fitting unit cells fit all the relevant parameters This flag tells the program to refine B factors of every batch from the very first cycles of refinement This is in contrast to the default procedure where the B factors are fit only after the convergen
39. mosaicity If there are too many preds you will have to decrease the mosaicity Let s say you started with a mosaicity of 0 4 degrees and you need to decrease it a bit Here s how to change it MOSAICI TY 0 3 GO Check again Let s assume you ve got it right now Note it If you have to go to unreasonable values of mosaicity gt gt 1 degree see Mosaicity and Spot Shape later in this manual Look at the integration boxes in the zoom window Zoom in enough to see the spots Are the preds big enough to cover the spot Are they too big Are the reflections so close together that the boxes overlap adjacent spots lots of red circles If you have overlaps that a small adjustment of the spot or box size is not going to fix then you will have to adjust your data collection perhaps by changing the oscillation range or the crystal to film distance Let s say that your reflections are smaller than the inner white jagged outline which is the Denzo spot area and that a slight adjustment in the box size will prevent most of your overlaps In the site dat file the spot size was defined as a 0 75 x 0 75 mm ellipse and the box was 3 5 mm on a side Change that now SPOT ELLIPTICAL 0 5 0 5 0 0 BOX 2 5 2 5 GO GO GO Hit Prof Fit R The display will show the profile fitting radius as a white circle and put preds only on those spots with I o greater than the weak level default value 50 If you don t have enough spots for profile fitting say
40. not FIT batch rotz because this parameter is very poorly defined by the intensities of observed partial reflections This is another very common mistake Unless you know what you are doing do not FIT crystal rotx roty rotz because if the spindle is even slightly misaligned the assumption that there is only one crystal orientation parameter for a large sweep of data will force incorrect restraints on the crystal orientation refinement E About fitting rotations Changes in rotations like crystal rotx roty and rotz are expressed as small rotations call them A about the laboratory frame of reference These A A Ay and A are used because to a first approximation they commute with one another commute means that the order in which they are applied is irrelevant This in turn is because these A are small typically less than one degree The crystal or cassette rotations Rx Ry and R on the other hand do not commute with one another because their values tend to be large much greater than one degree So when you ask Scalepack postrefinement or Denzo to fit these 125 The HKL Manual rotations what is actually happening is that the A are being refined After each refinement cycle the A are converted into changes in Rx Ry and R by a complicated algebraic relation Those of you with sharp eyes will have noticed in Denzo that the shifts reported by the program when fitting the crystal rotations do not correspond to the changes
41. not by the batch number format defaults example FILE value filename see also note below None FILE 101 frames scale lysoz x EP The FILE must come after FORMAT because the syntax depends on which input format is being modifiers read The FILE must not be followed by number in the case of archive denzo_york and denzo_york1 formats where the number after the word FI LE is not given because the batch numbers are already stored in the file If you want to change the batch numbers in these file formats see the ADD command described above Tells the program what parameters to fit in postrefinement and specifies the group of files over which the fitting is to be performed Postrefinement fit can be applied to an entire set of batches one batch being the entire set of frames for example using the modifier crystal or to each individual input file using the modifier film or batch crystal specifies that the fit operation is over the entire set of frames specified by a range and restrains the fit parameter to have exactly the same value over this range batch specifies that the fit operation is performed on each member of the set of frames specified by a range parameters format default examples THE SCALEPACK KEYWORD DESCRIPTIONS film alias for batch a b c unit cell lengths Values returned are real space alpha unit cell angles Values returned are real space beta gamma rotx crystal orientation
42. of reflections in a resolution shell used to define that resolution shell for the purpose of refining partiality Default is 100 LAMBDA REDUNDANT format default X ray wavelength Same as keyword WAVELENGTH wavelength in Angstroms 1 5418 Cu K 59 THE HKL MANUAL example LAMBDA 1 5418 LATTICE REDUNDANT Lattice type Note that this command is superseded by the SPACE GROUP command for most routine use format single letter or single letter with name Can also specify lattice type Possible choices are P C A B F IandR default P ericlinic example LATTICE P monoclinic LEFT MARGIN The left margin in mm from the left edge of useful data It is defined in the raw data convention not in the display program convention format LEFT MARGIN value_ mm default 0 0 example LEFT MARGIN 2 0 LIST This command is normally used in the manual mode It tells the program to provide a list of the current parameters This is convenient if you forget where you were in the refinement or you want to copy down the latest parameters To use just type LI ST LONGEST VECTOR In the autoindexing subroutine of Denzo LONGEST VECTOR denotes the longest real space vector in A the program expects to find in the reduced primitive unit cell If the autoindexing fails and you are confident of your X BEAM Y BEAM and DISTANCE values then this is the next parameter to try changing Typically you would either input a value c
43. or batches It adds a factor of scale1 scale2 scale restrain to the target function minimized in scaling This only applies to batches between which you add partials For very thin frames this is almost obligatory The value should roughly represent the expected relative change in scale factors between adjacent frames SCALE RESTRAIN value not turned on SCALE RESTRAIN 0 01 expect 1 change between adjacent frames Substitutes for the wildcard to specify a group of files to be read See FI LE keyword SECTOR integer to integer no default SECTOR 1 to 40 Sector width can be specified in degrees A pseudofilm is a sector width s worth of data from one detector Valid for area detector data only default value 5 degrees SECTOR WIDTH value 5 degrees SECTOR WIDTH 3 0 Cutoff for rejecting measurements on input Default 3 0 Be careful if you increase this What is the rationale for using o cutoff 3 0 in Scalepack Wouldn t you want to reject all negative intensities Why shouldn t you use a o cutoff 1 0 or zero The answer is as follows The best estimate of I may be negative due to background subtraction and background fluctuation Negative measurements typically represent random fluctuations in the detector s response to an X ray signal If a measurement is highly negative lt 30 than it may be more likely the result of a mistake rather than just random fluctuation If one eliminates negative fluctuations but not
44. parameters deduced in Denzo roty rotz wavelength incident wavelength Untested mosaicity mosaicity as defined in Denzo in degrees fit modifier parameterl filmnumber to filmnumber the default is that nothing is fit unless specified FHT crystal roty 1 to 137 FIT crystal mosaicity 1 to 5 7 to 10 102 to 104 to 137 FIT batch rotx 1 to 137 Most mistakes in Scalepack can be attributed to errors in FIT commands because the program is very sensitive to mistakes in the batch numbers If you input non existent batch numbers or define overlapping ranges e g 1 to 10 5 to 20 the program is likely to fail in a strange way If you specify a range of numbers the program will only use the batch numbers that exist within the range For example if your batch numbers go from 1 to 40 and 70 to 90 you can get away with saying say FIT batch parameter 1 to 90 which is the same aSFIT batch parameter 1 to 40 70 to 90 ForFIT crystal these two inputs are not equivalent In the case of FIT crystal parameter 1 to 90 one value will be fit for all batches In the case of FIT crystal parameter 1 to 40 70 to 90 two values will be fit one for each range Note that different parameters may be fit over different ranges and either over batch or crystal You can also mix batch and crystal for the same parameters E gt Do not fit unit cell parameters restrained by space group symmetry For example if you have space group P3 you must not fit b E Do
45. position of the direct X ray beam for each frame With automatic image plate scanners like the MacScience R axis or Mar this is generally not a problem because the detector and the beam do not move much during data collection and the scanning mechanism is accurate enough to scan the IP in a reproducible fashion Thus once the position of the direct beam is established for one frame it is pretty much the same for all the frames of the series Cassette based systems like the Fuji scanners found at several synchrotrons CHESS F1 Photon Factory Brookhaven X25 X4 etc do not have this reproducibility however This is due to the slight slop inherent in the positioning of the IP in the cassette the attachment of the cassette onto the camera carousel and the positioning of the IP in the scanner magazine Thus the x BEAM and y BEAM positions on each frame can vary beyond the normal radius of convergence for refinement in Denzo One way to overcome this is simply to examine each frame one by one and note down the position of the direct beam in each case and enter this value into the command file for batch processing Another way is to index the first frame and refine each subsequent frame interactively one by one after determining the direct beam position again by inspection of each image With a fast computer this is a viable albeit tedious option Note however that you should not keep re running the autoindexing routine This is because the au
46. potential derivative 98 SCENARIOS E you should pay most attention to the low resolution shells where the measurements are most accurate 1 If y 1 then your crystal scales well with the native so it is not a useful derivative Remember derivatives are different 2 If x very large say 50 then you have either severe non isomorphism sorry your have very heavy substitution but it is rare to have such a strong isomorphous difference signal or your derivative is misindexed relative to the native Reindexing is only a potential problem in the case of all trigonal P6 R3 P4 14 I23 and P23 space groups and in cases where there are accidental metric tensor symmetries e g monoclinic b 90 or orthorhombic a b To reindex apply an HKL MATRI X to test the other possibilities discussed in Scenario 5 and see which one gives the lowest y and Rmerge That is then the correct index 3 If y 10 and the detector is not a problem as indicated by R R then this indicates a potential derivative because you know that the differences are much larger than the expected measurement error 4 If 2 you either have a poorly substituted derivative or a slight detector problem Either way this is not going to be your best derivative The resolution shell where the x drops to about 2 represents the useful limit of your derivative measurements EP Instead of three 1 degree photographs in many cases you can get away wi
47. real image Exit from program Denzo Same as STOP or a semicolon This is one part of the alignment corrections needed for the spiral MacScience and MAR scanners It is the radial departure in mm of the scanning heads from perfect concentricity with the imaging plate A perfectly aligned scanner would have a RADI AL OFFSET of 0 0 The radial offset is perpendicular to the angular offset A diagram of these offsets is shown under the ANGULAR OFFSET keyword format default example RASTER format default example RAW DATA FILE ALPHABETICAL KEYWORD DESCRIPTIONS RADIAL OFFSET value_ mm 0 0 RADIAL OFFSET 0 43 Pixel size in the horizontal slow scanning dimension real number in mm defined by the values in the FORMAT statement RASTER 0 1 This defines the detector data file name template If the file name contains a run of characters then they will be substituted with the numerical value of the SECTOR argument This allows for easy processing of a large series of sequential images format RAW DATA FILE file name default none examples RAW DATA FILE bigdisk dir frames xtal00l osc RAW DATA FILE vms data dir xtal osc B in the first example the single quotes are necessary or else the directory names in square brackets will be ignored commented out E You must not repeat this command before the next GO statement RECORD LENGTH Defines the number of pixels in the data record It is equal to the number
48. rotx 1 to 160 FIT batch roty 1 to 160 FHT batch mosaicity L to L60 FORMAT denzo_ip HKL MATRIX 1 0O O reindexing matrix h h CVS 0 kI k OO also works for P312 P3121 P3221 P3112 and P3212 SECTOR 1 to 60 FILE 1 frames myxtal2 t x this data will be reindexed by matrix above Unit Matrix HKL MATRIX 1 0 0 data set 1 to be scaled with reindexed data O 1 Q OF 0 1 SECTOR 1 to 60 FILE 101 frames myxtall x this data will not be reindexed OUTPUT FILE myxtals12 sca output combines 2 data sets with consistent indexes 106 SCENARIOS P6 P4 or H3 Reindex data and combine with another data set NUMBER OF ZONES 10 number of resolution shells for statistics ESTIMATED ERROR 0 04 0 04 0 04 ESTIMATED ERROR for each resolution shell 0 04 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 4 multiplicative factor applied to input o REJECTION PROBABILITY 1 0e 4 WRITE REJECTION FILE 0 9 reject RESOLUTION 40 3 3 SPACE GROUP P6 REFERENCE BATCH 1 ADD PARTIALS 1 to 60 101 to 160 POSTREFINE 10 good idea to POSTREFINE because this will affect partials PET crystal am T to T60 Pl ve ry stad Gs 1 to 60 FIT batch rotx 1 to 160 FIT pacten roty Ito 160 FIT batch mosaicity 1 to 160 FORMAT denzo_ip HKL MATRIX O 1 O reindexing matrix h k i a kI h CS Ore elt 1 1 also works for p4 and h3 SECTOR 1 to 60 FILE 1 frames myxtal2 x this data will be reindexed by matrix
49. size of the crystal shape of the X ray beam etc Denzo allows you to change the size and shape of the pred spot to match the actual reflection better The way to judge the correct spot size is to zoom in on an area of the image and compare the magnified shape of the actual spot with the magnified pred shape You can t use different spot shapes for different regions of the image If your spot shape varies over the image you ll have to choose the best compromise This is not so bad though because you pay only a small penalty for including the background in the spot when using profile fitted intensities The keyword SPOT is modified by either radius or elliptical followed by numbers It is convenient to get used to using SPOT elliptical because a circular spot can be described by an ellipse whose major and minor axes are equal and you will undoubtedly run into elliptical spots one day A typical command for an elliptical spot would be SPOT elliptical 0 6 0 5 0 0 G0 where the first number is the length in mm of the major semi axis the second number is the length of the minor semi axis and the third number is the clockwise angle between a horizontal line and the major semi axis Once you have enlarged a region of your image it is quite easy to find the appropriate spot shape An idealized example is shown below pred SS minor semi axis spot x background border major semi axis Figure 4 DENZO spot terminology Note that most of the time
50. squares refinement process is used to improve the prediction The parameters describing the measurement process either have to be known a priori or have to be estimated from diffraction data by the refinement procedure Depending on the particulars of the experiment the same parameters e g crystal to detector distance are more precisely known a priori or are better estimated from the data Denzo allows for the choice of fixing or refining each of the parameters separately However using it well requires considerable knowledge of diffraction experiments Fortunately the FI T ALL option and detector specific default values seem to be reliable under most conditions Eigenvalue filtering Occasionally the refinement can be unstable because of a high correlation among some parameters High correlation makes it possible for the errors in one parameter to compensate partially for the errors in other parameters If the compensation is 100 the parameter would be undefined but the error compensation by other parameters would make the predicted pattern correct In such cases eigenvalue filtering described in Numerical Recipes as Singular Value Decomposition is employed to remove the most correlated components from the refinement and make it numerically stable Eigenvalue filtering works reliably when starting parameters are close to correct values but may fail to correct large errors in the input parameters Once the whole data set is integrated the global r
51. than preds increase the mosaicity more preds than observed reflections decrease the mosaicity Useful increments are 0 1 for typical protein crystals and 0 02 for virus crystals Depending on how dark you have set your display you can get fooled into thinking that your mosaicity is higher than it really is because you are trying to get preds to cover very weak reflections which result from diffuse or satellite scattering To get around this dim your display so that only the reflections representative of your Bragg scattering are displayed Then adjust the mosaicity as above until a good match is seen A third and more precise way of setting the mosaicity is to examine the histogram that Denzo produces in the log file or outputs to the screen if working interactively after the CALCULATE GO statement is given you can do this now but it is the last operation you will be able to execute on this image unless you restart Denzo since the SECTOR argument is incremented with the CALCULATE command and thus a new frame 1s read so hold off on this until you ve done the best you can with matching up preds to reflections and after you ve adjusted the spot shape and size The use of the histogram is discussed in the section Understanding the Denzo Log File later in this chapter To summarize briefly you want to adjust the mosaicity so that the observed average partiality histogram declines almost to zero in the last mosaicity zones It
52. the SCALE ANOMALOUS flag is used You can only use SCALE ANOMALOUS when you have enough redundancy to treat the F and F completely independently 113 The HKL Manual Scenario 8 Fitting polarization NUMBER OF ZONES 10 number of resolution shells for statistics ESTIMATED ERROR 0 04 0 04 0 04 0 04 ESTIMATED ERROR for each resolution shell 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 3 multiplicative factor applied to input o REJECTION PROBABILITY 0 0001 SPACE GROUP P422 OUTPUT FILE mydata sca REFERENCE BATCH 1 ADD PARTIALS 1 to 90 reject WRITE REJECTION FILE 0 9 commented out here because we do not want to increase the reject list FORMAT denzo_ip POLARI ZATION denzo 0 9 corrected POLARIZATION correction is made here 0 8 SECTOR 1 to 90 FILE 1 frames Denzo myxtal x Polarization correction testing is done in two stages First you scale the data as in Scenario 1 writing a rejection file Then second you run the command file above which changes the POLARI ZATI ON The log file is then examined to see if the x crystalhave decreased compared to the previous round when no polarization correction was done If the y have improved a significant improvement may be as low as 0 01 then you know that this is a better value for the polarization If there is no change in the y then you cannot establish whether the polarization was right or wrong This requires a tetragonal crystal trigona
53. the desired resolution limit Long unit cells large mosaicity or large oscillation range all affect spot separation and potential overlaps Some overlaps are immediately visible the ones arising from a long unit cell axis in the plane of the detector At high resolution due to the weakness of the spots the overlaps may be less obvious The simulation of a diffraction pattern based on indexing of the first image and proposed data collection protocol is the right tool to define an adequate oscillation range and correct detector placement There is no particular need to collect fully recorded reflections so the optimal oscillation range is typically narrow or even equal to a fraction of the crystal mosaicity Problems are best diagnosed by collecting data with benchmark high quality crystals e g tetragonal lysozyme There is no particular advantage of lysozyme crystals with the possible exception of how easy it is to grow them and a larger unit cell crystal would be preferable e g tetragonal chymotrypsin One should expect very high quality data from test crystals The resulting anomalous difference Fourier map should identify all the sulfurs in the protein The detector parameters refinement should produce a very small spread tens of microns hundredths of a degree from one image to another Such a test may require the mounting of a test crystal in a way that avoids slippage and minimizes absorption Rmerge Statistics in the range 2 3
54. the initial orientation is found all of the data can be included in the subsequent refinement There are two ways in Denzo to reject reflections corrupted by ice rings The elegant way exploits the fact that the distinguishing feature of spots corrupted by ice rings is a very high and irregular background Denzo assesses the quality of the background of each spot against criteria for slope and uniformity and if a certain fraction of the pixels in the background fail this test then the whole spot is rejected This is controlled by the keyword REJECT fraction The default value for this rejection is 0 75 1 e 75 of the background pixels must be acceptable If you have an ice ring you can increase REJECT fraction until most of the spots in the ice ring are no longer picked up Useful increments are 1 2 at a time Do this while you are looking at the display just as you did for spot shape and mosaicity As you increase the reject fraction the number of good preds in the ice ring should decrease while leaving most of the others alone The second way to avoid ice rings in your data is to simply process your frames twice once at a resolution cutoff below the ice ring and then again for the data beyond the ice ring The two data sets are then merged in Scalepack along the lines of Scenario 10 in The Scalepack Manual This is more of a brute force approach and is likely to result in the loss of more good data than the approach outlined above Re
55. the primitive choice i e H3 or H32 not R3 or R32 The SPACE GROUP must be declared before the FI T ce11 command can work E Denzo does not actually care about the SPACE GROUP per se It indexes films by Bravais lattice format default example and does not take into account systematic absences arising from two folds screws etc These are accounted for in Scalepack when the data are reduced So for the purposes of Denzo the resulting x file will be the same whether you declare the space group to be P222 or P212121 or P3 and P6122 SPACE GROUP modifier Pl SPACE GROUP P212121 69 THE HKL MANUAL SPINDLE AXIS ORIENTATION AXIS 1 The reciprocal lattice vector parallel to the spindle axis Same as keyword ORI ENTATION AXIS 3 integers representing the reciprocal vector 0 0 1 c axis SPINDLE AXIS 1 0 O a axis SPINDLE AXIS 1 O 1 anon major axis Defines the shape and area of the region in the box where the spot is to be found radius Defines the spot to be circular with a radius value in mm The center of the spot area and of the box coincide elliptical Defines the spot to be ellipsoidal with characteristic major and minor semi axis lengths in mm along with the clockwise angle between a horizontal line and the major semi axis radius value_ mm elliptical valuemajor valueminor_ mm valueangle_ degrees radius 0 35 elliptical 070 Uno 0 0 See diagram fig 4 in Mosaicity and Spot Shape s
56. then you should consider whether you have detector problems Likely culprits include the shutter and the spindle Note that bad crystals split twinned and ice rings can also produce this phenomenon as can the inclusion of a few frames of data from a badly decayed crystal If R R and your X are high then your crystal may have suffered radiation damage which cannot be corrected by scale and B factors This type of damage may not even increase B factors or decrease scale factors This is frequently seen with non frozen crystals The X is your best indication of whether your error model accurately reflects the actual errors in your data Since the errors in your data are expressed in the o value you attach to every I it is important that your error model be correct values of close to 1 indicate that your model is pretty good If your X values deviate much from 1 you can fix this by adjusting the value of the error scale factor Note that you don t have to hit 1 0 since X goes as the square of your error parameters so 20 deviations from 1 0 represent only a 10 adjustment to o which is OK for most practical purposes Even a X 2 is unlikely to seriously affect your subsequent calculations SCENARIOS Scenario 2 Scale Data from One Crystal not Processed with Denzo scalexengen com NUMBER OF ZONES 10 ESTIMATED ERROR 0 04 0 04 0 04 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 0 0 04 REJEC
57. you have called a FI T command Consider the following case you want to refine the crystal rotation values so you type FIT CRYSTAL rotx roty rotz Now let s say that you also want to enter a new value for the distance before giving a GO command You might think to try FIT CRYSTAL rotx roty rotz DISTANCE 200 GO but this won t work Program will not recognize the 200 and it will give you an error message plus it will try to refine the distance value as well The solution is to use the END FIT command as follows FIT CRYSTAL rotx roty rotz END FIT DISTANCE 200 GO The same goes for the END FIX command The option FIT cell when used in conjunction with the SPACE GROUP or LATTICE command tells the program to fit the unit cell lengths and angles The program is smart enough to know which lengths and angles are to be fit and which are to be fixed For example it knows enough to keep all angles at 90 degrees in orthorhombic space group Likewise it also knows enough to keep a b in tetragonal space groups etc The LATTI CE or SPACE GROUP must be declared before FIT CELL can be used In practice most often option used is FIT ali This will fit all parameters that can be sensibly fit given the circumstances space group detector Fitting all parameters is now stable even at low resolution all a t b 6 alpha beta gamma cell CRYSTAL rotz rotx roty CASSETTE rotz rotx roty radius DISTANCE X BEAM Y BEAM Y SCALE SKEW FILM rotatio
58. you need 10 numbers following the keyword estimated error one for each zone E The error estimates to do not all have to be the same The estimated error applies to the data which are read after this keyword so you can apply different error scale factor to subsequent batches by repeating this input with different values This is an important point if you enter data from a previous Scalepack output that does not need its o to be increased The error estimates should be approximately equal to the R factor in the table at the end of the output for resolution shells where statistical errors are small namely the earlier resolution shells where the data is strong This is a crude estimate of the systematic error to be multiplied by I and is usually invariant with resolution Default 0 06 i e 6 for all zones format ESTIMATED ERROR valuel value2 value3 value etc default each value set to 0 06 example ESTIMATED ERROR 0 02 0 03 0 03 0 03 0 04 0 04 6 zones EXTEND PARTIALS DO NOT EXTEND PARTIALS Some partially recorded reflections may be predicted by Denzo or Scalepack to start or end their Bragg condition in between consecutive frames due to small variations in crystal 123 The HKL Manual FILE FIT 124 format default orientation from frame to frame For these reflections only there are two choices of defining where the reflection started or ended including the extra frame or not EXTEND PARTIALS tells Scalepac
59. 0 0 2n 0 2n 0 0 0 2n 0 0 0 2n 0 0 2n 0 2n 0 E Note that for the pairs of similar candidate space groups followed by the or symbol scaling and merging of diffraction intensities cannot resolve which member of the possible pair of space groups your crystal form belongs to 119 The HKL Manual 120 THE SCALEPACK KEYWORD DESCRIPTIONS Alphabetical Scalepack Keyword Descriptions ADD format default example ADD PARTIALS modifier format default example Increments batch numbers by a constant to every batch from this point on until another add command is read Useful to make unique batch numbers from two or more files which have the same batch numbers inside For example Denzo_york1 format embeds the batch number in the x file ADD value nothing added ADD 1000 This will add 1000 to each batch number Tells the program to add partially recorded reflections among consecutive batches even if the batches do not have consecutive numbering Essentially obligatory all partials are added over all consecutive batches ADD PARTIALS starti to endl start2 to end2 etc do not ADD PARTIALS ADD PARTIALS O to 49 51 to 99 100 to 149 Be sure that ranges of numbers do not overlap ANOMALOUS format default example Flag for keeping Bijovets J and I separate in output file If the ANOMALOUS flag is on anomalous pairs are considered equivalent when calculating scale and B factor
60. 0 166551 0 942517 0 00067600 0 00903865 0 00142214 0 954577 0 021450 0 297193 osc start osc end dist in pixel units wavelength crystrotz roty rotx mosaicity 0 00000 1 50000 1332 822277 1 54180 72 1579 178 6024 170 2019 0 2500 these are the actual reflections now 1 2 3 4 5 6 7 8 9 10 11 12 13 27 7 5 1 249 8 231 0 1 82 20 4 0 919 29 6 684 4 0 172 1529 5 27 7 4 1 68 2 50 5 1 69 19 6 0 920 37 6 702 4 0 208 1119 9 23 5 I5 1 125 2 208 5 1 73 25 5 0 922 49 9 462 7 0 261 1412 3 22 5 171 306 9 301 4 1 65 35 9 0 919 53 8 414 9 0 354 1029 9 23 5 14 1 1413 6 1409 2 3 28 68 4 0 924 56 7 481 6 0 224 1397 3 some lines skipped T O 12 32 1 118141 5399 24 77 10031 0 997 1372 8 1196 3 0 191 3094 7 1 2 3 h k l not reduced 4 flag O full 1 partial 5 I by profile fitting 6 I by profile summation 7 x of profile fitting 8 o of I 9 cosine of incidence angle at detector 10 x coordinate of predicted centroid in pixel units 11 y coordinate of predicted centroid in pixel units 12 Lorentz polarization obliquity factor 13 strength of averaged profile arbitrary units In columns 5 6 and 8 the intensity of very strong reflections 1 e overflows will be written as an integer rather than as a floating point number see example on line T above Scalepack understands this Denzo convention 999 17 17 11111111111111111 11111111111111111 11111111111111111 11111111111111111 11111100200111111 11111022222011111 1111022222220111
61. 1 11110222222201111 11112222222221111 11110222222201111 44 999 flag for end of reflection list 17 17 size of the integration box left this is a saved spot profile for use in the display program if you want to overlay the final predictions on the original diffraction image The rectangle is the box 1 s represent the background area 0 s the guard area and 2 s the spot area REFINEMENT 11110222222201111 11111022222011111 11111100200111111 11111111111111111 11111111111111111 11111111111111111 11111111111111111 error density 0 750 these are the final Denzo parameters error positional 0 030 as refined from this image sector 1 raw data file alpha2 image useful for restarting data processing profile fitting radius 40 00 overlap spot monochromator 0 111 oscillation start 0 00 end 1 50 wavelength 1 54180 mosaicity 0 250 if clipping out may start from here resolution limits 100 0 3 8 spindle axis 0 0 1 vertical axis 1 0 0 Space group h32 unit cell 108 740 108 740 108 740 89 433 89 433 89 433 crystal rotx 170 202 roty 178 602 rotz 72 158 cassette rotx 0 16 roty 0 00 rotz 0 00 distance 199 92 x beam 89 194 y beam 89 582 film rotation 0 000 radial offset 0 202 angular offset 0 144 crossfire y 0 138 x 0 132 xy 0 000 45 THE AKL MANUAL Input Formats and Conventions 46 The input to Denzo consists of the digitized diffraction image a disk file of optical density readings from
62. 1 139 visualisation of the data 10 VMS OPERATING SYSTEM 22 77 140 147 148 154 W wavelength 33 Cu K 1 5418 59 LAMBDA Denzo keyword 59 WAVELENGTH Denzo keyword 72 WEAK LEVEL Denzo keyword 29 73 weak reflections 90 weak reflections 44 weaker pixels 20 Weissenberg method 6 53 143 workstations Alpha 147 DEC Alpha 13 147 148 SGI 77 147 SGI Indigo 13 SUN 77 147 WRITE BADDIES Scalepack keyword 139 WRITE NO PREDICTIONS Denzo keyword 73 WRITE PREDICTIONS Denzo keyword 73 WRITE REJECTION FILE Scalepack keyword 135 140 Write Print XdisplayF button 19 X X BEAM Denzo keyword 25 28 53 73 81 142 x beam Denzo log statement 45 XDisplayF general description 17 xengen urf file format 127 X plor 131 xtal x files 15 X terminal 77 Y Y BEAM Denzo keyword 25 28 53 73 81 y beam Denzo log statement 45 Y SCALE Denzo keyword 33 48 49 74 york output file 74 146 YORK OUTPUT FILE Denzo keyword 74 165 The HKL Manual Z Zero sigma 158 Zoom close 21 Zoom In XdisplayF button in Zoom Window 20 Zoom Out XdisplayF button in Zoom Window 20 Zoom Window XdisplayF button 19 a b c reciprocal unit cell vectors 126 a B and y reciprocal unit cell angles 126 x y weighted error measure 10 15 40 54 64 113 114 117 136 149 151 152 chi square modifier to keyword PRINT 64 166 x weighted error measure 30 X weighted error measure 95
63. 2 0 Unit cell parameters in either real or reciprocal space UNIT CELL a b c a B y or UNIT CELL a b c a B y Lengths in Angstroms or reciprocal A angles in degrees none UNIT CELL 120 92 120 92 131 65 90 90 90 If the space group has been declared the program is smart enough to know which cell lengths should be equivalent and which angles are variable and it will hold those lengths equal or fixed throughout the refinement Tells the program to use the beam spot in the refinement This keyword has the opposite effect of the keyword NO BEAM This is useful in the initial steps of the refinement when not too many spots are matched by preds only Tells the program to use only the beam spot in the refinement Has the 71 THE HKL MANUAL effect of turning off the RESOLUTION LI MI TS keyword format USE BEAM USE BEAM only default not the default option See NO BEAM USE FIDUCIALS USE NO FIDUCIALS Tells the program to use the fiducials in the refinement Turned off with USE NO FI DUCI ALS Fiducial positions are entered with the keyword FI DUCI ALS modifier Oniy Tells the program to use only the fiducial positions in the refinement format USE FIDUCIALS USE FIDUCIALS only default not the default Default is USE NO FIDUCIALS USE PARTIALS POSITION USE PARTIALS NO POSITION Normally data reduction programs use only fully recorded reflections in the refinement of the crystal and detector positional parameters
64. 3 FILE 36 my_crystal_hi x SECTOR 184 to 238 FILE 184 my_crystal_hi x INCLUDE OVERLOADS SECTOR 244 to 283 FILE 1000 my_crystal_lo x OUTPUT FILE my crystalhilo sca number of resolution shells for statistics ESTI MATED ERROR for each res shell Set to zero because native has already been scaled good for CCD at CHESS expected fraction of outliers in the data one res range for both high amp low res use the first frame as reference for scaling reject outliers from previous rounds 20 cycles of postrefinement note that there are missing batch numbers created by the file statement below but that we do not need to mention this for postrefinement note that the consecutive sequences must be declared here otherwise program will try to ADD PARTI ALS from non consecutive frames High Resolution Data for the high res data long exposure the x files not numbered consecutively Note renumbering of sector 39 x to batch 36 frame 39 x is next SECTOR after 35 x because 36 37 38 were calibration frames there was change in crystal orientation at 184 x Note that now we re even with batch numbering Low Resolution Data now we go to low exposure low resolution data we put low resolution sectors in batches starting with 1000 just to keep it straight 115 The HKL Manual Scenario 10 Thin or few frames SCALE RESTRAIN And B RESTRAI N scalethins com 116 NUMBER OF ZONES ESTIMATED ERROR Occa
65. 60 101 to 160 201 to 260 301 to 360 POSTREFINE 10 good idea to POSTREFI NE because this will affect partials FIT crystal a 1 to 360 PII Crystal se L tO 360 PIT Dateh rotz Wh Veo 3 60 FLT batch roty L to 360 FIT batch mosaicity 1 to 360 FORMAT denzo_ip HKL MATRIX 1 0O O reindexing first matrix h h ET kik OROT SECTOR 1 to 60 FILE 1 frames myxtall x this will be reindexed by matrix 1 above HKL MATRIX 0 1 O data set 2 to be scaled with reindexed data EIN OE OO al SECTOR 1 to 60 FILE 101 frames myxtal2 x this data will be reindexed by matrix 2 above HKL MATRIX O 1 O data set 3 to be scaled with reindexed data ep Ge OTOS SECTOR 1 to 60 FILE 201 frames myxtal3 x this data will be reindexed by matrix 3 above HKL MATRIX 1 0 0 original unit matrix ODO OF ek SECTOR 1 to 60 FILE 301 frames myxtal4 x this is the reference index data set OUTPUT FILE myxtals12 sca output combines 2 data sets with same index SCENARIOS Scenario 6 Comparing two data sets from Scalepack sca lenatder com NUMBER OF ZONES 10 number of resolution shells for statistics ESTIMATED ERROR 0 0 0 0 Set to zero because native amp derivative have already been 0 0 0 0 0 0 0 0 0 0 0 0 assigned proper errors 0 0 0 0 ERROR SCALE FACTOR 1 0 don t need here because data already scaled SPACE GROUP 1422 SPACE GROUP of native amp derivative optional FORMAT Scalepack data has been processed already and is in Scalepack F OR
66. 7 parameters 6 25 28 34 78 FORMAT Denzo keyword 58 precession method 6 problems 8 9 19 96 113 135 151 quality 89 R axis 9 21 37 42 47 49 53 55 63 81 90 127 142 144 148 153 156 saturation 9 39 156 OVERLOAD VALUE Denzo keyword 63 shutter 8 96 151 single pixel spikes 9 spindle 96 spindle motor backslash 8 spiral 6 9 69 unusual instruments 42 vibration 17 Weissenberg method 6 53 56 Diff Vec 21 diffuse or satellite scattering 35 diffuse scattering 68 Dim XdisplayF button 20 DIP MacScience gt detectors 55 Display 151 DI STANCE Denzo keyword 28 33 53 54 distortion 24 26 27 30 144 DO NOT EXTEND PARTIALS Scalepack keyword 123 DO NOT REJ ECT OUTLIERS Scalepack keyword 123 E eigenvalue filter 34 88 90 95 97 134 Eigenvalue filter EIGENVALUE FILTER Denzo keyword 54 END FIT Denzo keyword 57 INDEX END FIT Scalepack keyword 123 END FIX Denzo keyword 57 END OF 0B Denzo keyword 42 54 153 END OF PACK Denzo keyword 54 error ERROR Denzo keyword 54 error model 10 34 40 54 89 96 113 error scale factor 87 96 123 ERROR SCALE FACTOR Scalepack keyword 95 97 123 estimated error 35 64 87 95 ESTIMATED ERROR Scalepack keyword 95 97 98 100 123 131 152 expected error 40 55 89 135 in the fitting 40 overload 55 partiality 55 positional 55 profile fitting 148 systematic error 55 95 123 y x factor 55 X weighted measure 54 ER
67. 8 X BEAM Denzo keyword 25 28 53 73 81 x beam Denzo log statement 45 Y BEAM Denzo keyword 25 28 53 73 81 y beam Denzo log statement 45 Bijvoet differences 48 74 136 box integration box 20 44 49 51 63 151 Box integration box 145 BOX Denzo keyword 14 25 29 42 52 58 143 154 Bravais lattice 24 27 69 117 118 144 centered monoclinic 119 centered orthorhombic 119 centered tetragonal 119 cubic 27 74 114 118 hexagonal 27 74 118 monoclinic 27 orthorhombic 27 primitive monoclinic 119 primitive orthorhombic 119 primitive tetragonal 119 primitive triclinic 119 rhombohedral 27 118 tetragonal 27 74 114 triclinic 27 trigonal 74 114 C c weighted error measure 158 CALCULATE Denzo keyword 41 52 54 66 70 157 calculate Denzo log statement 42 CASSETTE Denzo keyword 25 28 33 42 CASSETTE Denzo keyword 52 CCD detectors 6 9 69 74 115 145 CCP4 file format 146 160 CELL Denzo keyword 53 cell reduction 24 centered space group 78 chymotrypsin 8 circle 59 Close XdisplayF button 20 Color XdisplayF button 19 command file 13 46 auto dat 26 batchDenzo com 41 experiment dat example 26 findbeam dat 42 refine dat 4l scaletwosca com 101 Site dat example 25 Computer problems and questions 146 computer recommendations 77 Control input in Scalepack 88 conventions detector orientation 47 Gravity beam orientation system 4
68. 8 input to Denzo 46 International Tables 117 rotation in Denzo 79 Spindle beam orientation system 48 COUPLI NG Denzo keyword 53 CROSSFIRE 33 53 154 CROSSFIRE Denzo keyword 53 CRYSTAL Denzo keyword 54 79 CRYSTAL rotz roty and rotx Denzo keyword rotation conventions 80 crystal symmetry Bravais lattice 24 27 69 117 118 144 pseudo symmetry 24 crystallography as an iterative process 10 crystals cracked crystal 29 frozen 39 intrinsic parameters 6 orientation 23 26 parameters 32 satellite crystals 9 25 29 135 size 36 slippage 8 twinned crystals 29 31 146 Crystals twinned crystals 158 cutoff 68 D DEC Alpha 147 148 DEFAULT B FACTOR Scalepack keyword 122 DEFAULT SCALE Scalepack keyword 122 129 density of the spots 66 Denzo keyword modifier absorption of the film 56 Denzo_ip format 123 127 128 133 139 Denzo_york1 format 121 124 127 128 133 139 146 detectors 8 9 alignment 9 55 calibration 8 CCD 6 9 69 74 115 145 data file template RAW DATA FILE Denzo keyword 67 DIP MacScience 47 49 52 55 63 81 distortion 9 22 dynamic range 19 Fuji 47 Hamlin SDMS 122 147 archive file format 122 124 126 127 147 FIXED WINDOW Scalepack keyword 126 Image Plate 5 9 Kodak 47 manual IP 9 28 42 81 Mar 9 47 49 63 81 MAR 37 multi wire proportional counters 9 non planarity of film or IP 9 orientation 23 orientation conventions 4
69. 8 85 24 147 65 primitive orthorhombic 0 30 41 07 65 26 124 78 90 00 90 00 90 00 41 07 65 26 124 78 90 18 90 58 89 40 C centered orthorhombic 2 27 41 07 252 50 65 26 90 00 90 00 90 00 41 07 252 50 65 26 89 92 90 60 81 22 I centered orthorhombic 4 72 41 07 77 48 269 93 90 00 90 00 90 00 A107 TT 48 2609 93 C2077 T299 Blase F centered orthorhombic 3 33 41 07 136 43 252 50 90 00 90 00 90 00 41 07 136 43 252 50 92 71 98 78 73 08 primitive monoclinic 0 18 41 07 65 26 124 78 90 00 90 58 90 00 41 07 65 26 124 78 90 18 90 58 89 40 C centered monoclinic 2 21 41 07 252 50 65 26 90 00 90 60 90 00 41 07 252 50 65 26 89 92 90 60 81 22 primitive triclinic 0 00 41 07 65 26 124 78 89 82 69 42 89 40 autoindex unit cell 41 07 65 26 124 77 90 00 90 00 90 00 crystal rotx roty rotz 131 740 39 042 4 743 Volume of the primitive cell 334451 Autoindex Xbeam Ybeam 96 42 95 58 crystal rotx 48 260 roty 39 042 rotz 175 257 crystal Protx 48 260 roty 39 042 rotz 4 743 crystal rotx 131 740 roty 39 042 rotz 4 743 crystal rotx 131 740 roty 39 042 rotz 4 743 Blue lines represent the 14 possible Bravais lattices in decreasing order of symmetry with primitive cubic being the highest symmetry and primitive triclinic the lowest After the lattice name is a percent value which represents the amount of distortion the unit cell parameters would suffer in order to fit the lattice Next to this percentage are the distorted to fit unit c
70. 94 94 95 97 98 100 102 111 112 114 115 116 117 121 141 142 142 142 143 145 145 146 148 149 151 151 153 153 154 154 156 156 158 158 159 INTRODUCTION Preface The HKL suite is a package of programs intended for the analysis of X ray diffraction data collected from single crystals and consists of three programs XdisplayF for visualization of the diffraction pattern Denzo for data reduction and integration and Scalepack for merging and scaling of the intensities obtained by Denzo or other programs Copyrights The first version of Denzo and the first prototype of Scalepack were written by Zbyszek Otwinowski at the University of Chicago in 1983 Programs Denzo and Scalepack are 1985 by Zbyszek Otwinowski and 1997 by the University of Texas Southwestern Medical Center at Dallas The program XdisplayF was developed by Wladek Minor and is 1991 by the Purdue Research Foundation and 1997 by the University of Virginia Patent Foundation The HKL Manual Edition 4 also known as The Denzo Manual and The MacDenzo Manual was written by Daniel Gewirth with the cooperation of Zbyszek Otwinowski and Wladek Minor and is 1993 95 by Daniel Gewirth Edition 5 was revised and updated by Zbyszek Otwinowski and Wladek Majewski The new typographic design was created by Grzegorz Jablonski The subsequent releases were further corrected and improved Some new material was incorporated in response to que
71. ATED ERROR 0 04 0 04 0 04 0 04 ESTIMATED ERROR for each resolution shell 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 4 REJECTION PROBABILITY 0 0001 WRITE REJECTION FILE 0 9 reject RESOLUTION 40 3 3 SPACE GROUP p21 REFERENCE BATCH batch 1 ADD PARTIALS partials 1 to 180 POSTREFINE 10 FIT crystal a 1 to 18 FIT crystal b 1 to 180 FIT crystal c 1 to 180 Pld crystal beta 1 to 180 FIT batch rotx 1 to 180 FIT batch roty 1 to 180 FIT batch mosaicity 1 to 180 FORMAT denzo_ip HKL MATRIX O O 1 Uri eC L 0 e SECTOR 1 to 180 FILE 1 frames myxtal x OUTPUT FILE myxtal sca multiplicative factor applied to input o POSTREFI NE will affect partials reindexing matrix h J 1 this is 1 to keep determinant 1 I h OUTPUT FILE contains reindexed reflections SCENARIOS Change order of a b cto b c ain orthorhombic NUMBER OF ZONES 10 number of resolution shells for statistics ESTIMATED ERROR 0 04 0 04 0 04 0 04 ESTIMATED ERROR for each resolution shell 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 4 multiplicative factor applied to input o REJECTION PROBABILITY 0 0001 WRITE REJECTION FILE 0 9 reject RESOLUTION 40 3 3 SPACE GROUP P21212 can also use for P2221 REFERENCE BATCH 1 ADD PARTIALS 1 to 180 POSTREFI NE 10 good idea to postrefine independent of reindexing FHT crystal a 1 to 180 HT crystal ba 1 to 190 FIT crystal c 1 to 180 Pit baten rotx 1 to 180 FIT
72. Cassette and Off line Image Plate Users Detector Specifications TABLE OF CONTENTS CONN A U m p lt 13 16 17 22 22 23 23 32 35 37 39 40 41 42 44 46 47 47 48 49 50 75 75 76 77 78 81 82 THE HKL MANUAL The SCALEPACK Manual Program Description Statistics and Scalepack Implementation Limits on variable dimensions Platforms Scenarios List of Scenarios Scenario 1 Scale Data from One Crystal Processed with Denzo Scenario 2 Scale Data from One Crystal not Processed with Denzo Scenario 3 Heavy Atom Search Scale native plus a few frames of a potential derivative Scenario 4 Scale Multiple Native Data SetsTogether Data from Denzo Scenario 5 Reindexing Scenario 6 Comparing two data sets from Scalepack Scenario 7 Compare I and I detection of anomalous signal Scenario 8 Fitting polarization Scenario 9 High resolution low resolution data sets from one crystal Scenario 10 Thin or few frames SCALE RESTRAIN andB RESTRAIN Scenario 11 Space Group Identification Alphabetical Scalepack Keyword Descriptions HKLO amp A 20 Absorption Correction Factors Auto editing Autoindexing Box definition CCP4 Scaling Computer problems and questions Consequences of Profile fitting errors in Denzo DENZO output Display Example of Bad Detector Pathology Ignore Mosaicity Negative Intensivities Other Overload Refinement Rhombohedral Twinned Crystals Index 87 88 89 93 93 93
73. D FIT END OF JOB END OF PACK ERROR density overload partiality positional systematic V x factor FI DUCI ALS FILM absorption length output file rotation width FIT FIX a b c alpha beta gamma crystal rotx roty rotz cassette rotx roty rotz distance x beam y beam y scale skew film rotation 50 radius crossfire x y xy all FORMAT see Detector Specific Info GO BOX GNORE MAGI NG PLATE INPUT INTENSITY AVERAGING WINDOW LAMBDA LATTICE LEFT MARGIN LIST LONGEST VECTOR MONOCHROMATOR filter graphite MOSAI CITY NO BEAM NUMBER OF RECORDS OMI T radius elliptical ORIENTATION AXIS 1 ORIENTATION AXIS 2 OSCILLATION start end range step OVERLAP spot guard none OVERLOAD VALUE PLATE ROTATION PEAK SEARCH FILE peaks file POLARI ZATI ON PRI NT chi square no chi square fitted hkl memory no memory profiles no profiles shifts no shifts solver no solver Statistics no statistics PROFILE FITTING RADIUS QUI T RADIAL OFFSET RASTER RAW DATA FILE RECORD LENGTH REFINE PARTIALITY REFINE NO PARTIALITY REJECT fraction cutoff slope systematic RESOLUTION LIMITS SECTOR SKEW SPACE GROUP SPINDLE AXIS SPOT radius elliptical START REFI NEMENT STOP SWAP BYTES obsolete TITLE TOP MARGIN UNIT CELL USE PARTIALS POSITION USE BEAM USE FIDUCIALS USE NO FIDUCIALS USE PARTIALS POSITION USE PARTIALS NO POSITION VERTI CAL AXIS WAVELENGTH WEAK LEVEL WRI TE PREDI CTI ONS WRI TE NO PR
74. E BATCH in scaling procedure SCALE RESTRAIN 0 05 scale and b factors between adjacent frames cannot vary by more B RESTRAIN 0 1 than these values ADD PARTIALS 1 to 90 must have this for restraints to work number of cycles POSTREFI NE 10 of postrefinement FIT crystal a 1 to 90 FIT crystal b 1 to 90 refine UNIT CELL parameters based on all data from batches 1 FIT crystal c 1 to 90 through 90 FIT crystal beta 1 to 90 Space group dependent FIT patch rotx 1 to 90 crystal orientation of spindle axis angle will be refined for each FIT batch roty 1 to 90 batch frame separately FIT crystal mosaicity 1 to 90 one mosaicity value will be refined FORMAT denzo_ip SECTOR 1 to 90 FORMAT of the input intensity data these values will be substituted in for the Batch numbers FILE 1 frames Denzo myxtal x will start at 1 for file myxtal001 x and increase to 90 for file myxtal090 x It is not correct to try to bin the individual frames into larger batches to try to overcome the problem of few intersections between frames This is because you then lose the ability to ADD PARTIALS between the new bins You can however overcome the problem in another way by including instead the statement NUMBER OF I TERATI ONS Q In this case the data will not be scaled but simply merged Obviously this has its drawbacks SCENARIOS Scenario 11 Space Group Identification Scalepack can be used to determine the space group of your crystal
75. E GROUP ALPHABETICAL KEYWORD DESCRIPTIONS RES 2030 245 This is the counter and range delimiter for a series of frames when processing in batch mode The number in the sector argument is substituted for the or or variable in the RAW DATA FILE and FILM output file commands Make sure there are enough marks to accommodate all of the digits in the sector argument e g if you have sector values in the 100 s then there must be at least three marks in the file names specified in RAW DATA FILE and FILM output file The sector value can start at any non negative integer and is incremented by 1 every time an END OF PACK statement is reached The first number in the sector argument is the first frame and the last number is the last frame See also START REFINEMENT SECTOR integer to integer none SECTOR 1 to 50 SKEW refers to the non orthogonality of the vertical and horizontal scanning directions measured in radians For a perfectly orthogonal system SKEW is defined to be zero Generally SKEW values will be very close to this number It can be fitted in the later rounds of parameter refinement SKEW value_ dimensionless 0 0000 SKEW 0 00169 value taken from the erlier refinement Spiral scanners and CCD detectors should use the default value of zero which should not be refined The space group symbol or number as given in the International Tables Note that for rhombohedral space groups autoindexing only works for
76. E bar Use GNORE quadrilateral instead _ We re sure the mosaic spread is increasing How do we refine a slowly varying mosaic spread in Scalepack FIT batch mosaicity 1 to 60 mosaicity will be refined separately in each batch of data Orff lil crystal mosalerty 12 or 10 TI te 220 21 630 31 to 40 41 e250 5L to 60 mosaicity will be refined separately for 6 groups of batches with separate values for first second third etc groups of 10 batches These groupings may be different for mosaicity and for other unit cell parameters Q How do we handle a slowly varying mosaic spread in Denzo If you have 60 frames and you want to process first 10 with mosaicity 0 2 degree next 10 with mosaicity of 0 25 degree and so on up to the last 10 with mosaicity of 0 45 you may do that in the following way In the command file type MOSAICITY 0 2 SECTOR 1 to 10 GO to execute autoindexing refine dat refine first 10 images MOSAI CITY 0 25 SECTOR 11 to 20 refine next 20 images MOSAICITY 0 3 SECTOR 21 to 30 35 SECTOR 31 to 40 4 SECTOR 41 to 50 45 SECTOR 51 to 60 refine dat 0 refine dat refine next 20 images MOSAI CITY O refine dat refine next 20 images MOSAICITY 0 refine next 20 images MOSAI CITY O refine dat SS T m m m mm refine dat refine next 20 images END OF JOB Q Looking at the raw data I found that Denzo had some reflections spread over 3 or 4 frames Given the frame size and mosaic spread
77. EDI CTI ONS X BEAM Y BEAM Y SCALE YORK OUTPUT FILE ALPHABETICAL KEYWORD DESCRIPTIONS AIR ABSORPTION LENGTH Length of air or helium in mm over which the scattered X rays are attenuated by a factor of e 2 718 format AIR ABSORPTION LENGTH value_ mm default AIR ABSORPTION LENGTH 860 example AIR ABSORPTION LENGTH 200 for 2 5 A X rays ANGULAR OFFSET This is one part of the alignment correction needed for the spiral MacScience and MAR scanners It is the distance in mm that the reading head is offset in the non radial scanning direction from the center of the imaging plate See also RADI AL OFFSET format ANGULAR OFFSET value_ mm default ANGULAR OFFSET 0 0 example ANGULAR OFFSET 1 2 radial offset BACKGROUND Defines the shape and size of the background area where the background measurement takes place Usually defined to be outside of the spot area However when no background is specified the default area is the area just outside of the spot border modifiers radius defines the background to be circular with a radius value in mm The center of the background area and of the box coincide elliptical defines the background to be ellipsoidal with characteristic major and minor semi axis lengths in mm along with the clockwise angle between a horizontal line and the major semi axis Refer to the diagram under the keyword SPOT formats BACKGROUND radius value_ mm BACKGROUND elliptical value major_ mm value
78. EL you can exclude weak spurious peaks then run autoindexing and examine the superimposition of the autoindexing preds green circles on the image If most of them cover real reflections and not noise then you are succeeding sizes may have to be adjusted if too many spots are being removed by the overlap criteria Autoindexing does not use spots which Denzo determines are overlaps Finally you can try changing the number of peaks you select in the peak picking routine However since Denzo edits the list of peaks anyway this is unlikely to yield much improvement if the suggestions above do not work In that case do you have a better crystal 29 THE HKL MANUAL Misprediction a misindexing Kind of failed Predicted positions 30 results in location of reflection masks not corresponding to the positions of the Bragg peaks It is visibly obvious and is disastrous It may be due to forcing wrong space group symmetry misindexing or serious detector malfunction Autoindexing is based on the assumption that the diffraction spots are correctly mapped from detector coordinates to diffraction reciprocal space The origin of the diffraction space is defined by the position of the direct beam on the detector A substantial error in the beam position can shift the indexing of the diffraction pattern by an integer vector Such misindexing can be totally self consistent until the stage when symmetry related reflections are compared
79. HE HKL MANUAL 92 single zone the Rmerge 1n the final shell would be pretty good Rmerge overall when in fact it was substantially worse It is more sensible to divide your zones into equal volumes and have enough of them so that you can accurately monitor the decay with resolution 4 Omitting partially recorded reflections This has the effect of a reducing the redundancy and b eliminating poorer reflections Partially recorded reflections will always have a higher o associated with them because they have a higher total background due to the inclusion of background from more than one frame in the reflection 5 Scaling I and I reflections separately in the absence of a legitimate anomalous signal SCALE ANOMALOUS This has the effect of reducing the redundancy 6 Ignoring overloaded reflections using the GNORE OVERLOADS in Scalepack The intensity of overloaded or saturated reflections cannot be directly measured because obviously some of the pixels are saturated Profile fitting only measures these reflections indirectly by fitting a calculated profile to the spot using the information contained in the wings or tail of the spot Ignoring the inaccuracies inherent in this method by ignoring overloads may have a dramatic effect on Riepe mP GNORE OVERLOADS is often a useful tool however For example when calculating anomalous differences you do not want to use overloaded reflections because you are looking for very very small
80. I m puzzled by this They were near the spindle however so is this a Lorentz effect i e what is your criterion for excluding reflections with large Lorentz correction All reflections that completely cross Ewald sphere are included 153 The HKL Manual Q Do you only do the postrefine business on the final run after rejecting the rejects or do you do it right away Is there an isotropic mosaicity I can refine should I be refining in all directions or should I really be refining just mosxx as was in the sample file somebody handed me Does it matter as long as I m not going to re Denzo the frames I set it to 4 in Denzo but it seems to refine to around 25 or so so it seems pretty safe to me You can only refine isotropic mosaicity It is better to have larger mosaicity in Denzo and refined by Scalepack as Scalepack can reliable observed partial reflections as non diffracting but not the opposite If mosaicity is decreased some partial reflections will be relabeled as fully recorded Q In Scalepack I can postrefine all 6 mosaicity parameters but in Denzo I can only use one mosaicity parameter So how can I tell Denzo that the crystal was more mosaic in the y direction than the x Do I change the value of CROSSFIRE CROSSFIRE has to do with focusing of X rays not reflection width in spindle angle Both Scalepack and Denzo use isotropic mosaicity only Anisotropic option is not yet implemented Negative Intensivities Q Wh
81. It is entirely possible that both of the two data sets would use the same reindexing matrix to arrive at a consistent R3 index but the example below assumes that you have already determined or are just testing that the two H3 data sets have indices which are inconsistent with each other NUMBER OF ZONES 10 number of resolution shells for statistics ESTIMATED ERROR 0 04 0 04 0 04 ESTIMATED ERROR for each resolution shell 0 04 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 4 multiplicative factor applied to input o REJECTION PROBABILITY 1 0e 4 don t reject until you know reindexing makes the two data set WRITE REJECTION FILE 0 9 consistent reject RESOLUTION 40 3 3 SPACE GROUP R3 H3 in Denzo converted to R3 in Scalepack by reindexing matrix REFERENCE BATCH 1 ADD PARTIALS 1 to 90 101 to 190 POSTREFINE 10 good idea to POSTREFI NE because this will affect partials FIT rvs eds ax ule gO T90 BT eryseet 1 I to a 90 FT Darch rors Tto 190 ELT pateh roty 1 6 190 FIT batch mosaicity 1 to 190 FORMAT denzo_ip HKL MATRIX 1 O 1 reindexing first matrix I ADO Le al SECTOR 1 to 90 FILE 1 frames myxtall x this will be reindexed by matrix above HKL MATRIX 1 1 O data set 2 to be scaled with reindexed data SO ers S SECTOR 1 to 90 FILE 101 frames myxtal2 x this data will be reindexed by matrix 2 above OUTPUT FILE myxtals12 sca output combines 2 data sets with consistent indexing 109
82. LOAD VALUE defines detector saturation Systematic error in Denzo is a number that describes expected correlation between positional errors on the detector Outlier rejection happens when symmetry related reflections do not agree in measured intensity Q What exactly is the scanner doing to give such funny data Another problem with MAR scanners is sometimes a wrong scale factor applied to strong fluorescence signal compared to weak signal Setting OVERLOAD VALUE to below transition point between strong and weak range of data may help but this would eliminate a lot of reflections Transition point is roughly at about 10000 MAR units Q SHOW OVERLOADS does not work with most file formats It works now with all formats Refinement Q I have set up a batch job for a set of images taken from one crystal at CHESS I indexed the first one with autoindex and wrote a shell script I understand that with the Fuji scanner you 156 HKLQ amp A can give SECTOR I to n instead of giving the line refine n times But then some images did not refine because the beam position shifts a lot I picked the beam position in the display and wrote the following SECTOR 1 refine X BEAM Y BEAM refine So I still have one refine per image but can alter the beam position before the next image Is this the best way to do it it works or is there a better way This will work Autoindexing gives almost correct cell constants but the predictions are so
83. NIT CELL example RECSQ 0 0001 0 O 0 0004 0O 0 0009 to describe unit cell 100 50 33 90 90 90 REFERENCE BATCH Specifies which batch or film or set of batches or films will be the reference for the scaling and B refinement The scale and B factor for these are not refined More than one film or batch may be used as the reference This is important only for crystals which decay during data collection If the crystal is frozen and does not decay then the default may be used which is to let the eigenvalue filter define the overall scale and B factor With a large number of batches reliance on the eigenvalue filter is a little bit dangerous so you should consider using a reference batch number in those cases This keyword is entirely equivalent to the keywords REFERENCE BATCHES REFERENCE FILM AND REFERENCE FILMS The others exist because some people have a gramatical hangup about using the singular to describe more than one object format REFERENCE BATCH integer integer integer etc default no reference batch Eigenvalue filter defines overall scale and B example REFERENCE BATCH 1 3 4 5 REFERENCE BATCHES FILM FILMS Same as reference batch REJECT HKL Flag which tells the program to reject a list of individual h k l s which follow This is useful for iterative rejection cycles since the file containing the rejected reflections can be reread Each record after the title contains one reflection with the variables h k orig
84. NT statistics PROFILE FITTING RADIUS 20 0 ous RADIAL OFFSET 0 0 E RADI US Flat oo SPINDLE AXIS 064 5 0 VERTICAL AXIS 10 O 0 0 WAVELENGTH 1 5418 aTi WEAK LEVEL Bad full file size X BEAM FILM width 2 1 5418 Y BEAM FILM length 2 Format Specific Keywords Theirs default values can be found in Detector Specific Information section ERROR density FILM absorption FILM length Keywords with no defaults FILM width NUMBER OF RECORDS OVERLOAD VALUE lt without which the program absolutely will not run BOX FORMAT MOSAI CITY lt required for interactive processing F T lt required for batch processing FIT GO CALCULATE END OF J OB OSCILLATION or PRECESSION OVERLAP RAW DATA FILE GO END OF PACK FILM OUTPUT FILE NPUT lt which may or may not be required depending on the circumstances FI DUCI ALS GNORE OMI T PRINT fitted 76 PRINT hkl PRINT profiles PRINT solver READ AMATRI X RASTER RECORD LENGHT Y SCALE RESOLUTION LIMITS SPOT UNIT CELL SECTOR TITLE READ UMAT UNIX 1 0 LONGEST VECTOR RECOMMENDED HARDWARE Recommended Hardware The operative philosophy behind the programs Denzo XdisplayF and Scalepack is that computer hardware is cheap compared to all of the other expenses of crystallography Modern platforms are supported old ones are gradually rendered obsolete by their speed discontinuation or by other hardware requirements like memory Here s a list of curren
85. O If one writes york format files with Denzo what is the appropriate keyword to use with Scalepack I thought Denzo_york worked but it seems to fail The error is ERROR Unable 145 The HKL Manual to open I for reading but if the format is changed to Denzo _ ip the file is opened correctly but of course the data format is incorrect Use Denzo_york1 instead Scalepack will read also concatenated Denzo_york1 format but not Denzo _ip Denzo _york is another creation by Eleanor Dodson that hopefully died by now The error is ERROR Unable to open 1 for reading comes from the fact that Denzo _york and Denzo _york1 do not require staring batch number as it is coded in the header of Denzo _york formats The 1 is interpreted as file name with hkl s instead and Scalepack cannot open it if unless you called output from Denzo 1 To go to CCP4 format you should use different option in Denzo YORK OUTPUT FILE That will produce somewhat different x file that can go into rotaprep part of ccp4 package Lack of satisfactory result from Scalepack maybe either from misunderstanding how the program works or from problems in the data collection Scalepack algorithms are fairly similar to Rotavata Agrovata and should not give qualitatively different results I am still unable to go to SCALA since the ROT information does not carry through rotaprep Any advice would be appreciated So far though Scalepack does the same as rotavata and agrova
86. ON LENGTH Denzo keyword 51 Alpha DEC workstation 147 148 ANGULAR OFFSET Denzo keyword 33 51 66 angular offset in Denzo output file 45 anomalous signal 8 88 91 ANOMALOUS Scalepack keyword flag 112 121 change of anomalous signal sign 102 NO ANOMALOUS Scalepack keyword 129 NO MERGE original index Scalepack keyword 130 OUTPUT ANOMALOUS Scalepack keyword 132 Patterson difference 113 SCALE ANOMALOUS Scalepack flag 91 113 136 Scalepack scenario 112 archive file format 124 127 147 BACKGROUND CORRECTION Scalepack keyword 122 FIXED WINDOW Scalepack keyword 126 Archive file format 139 ARCHIVE file format NO PROFILE TEST Scalepack keyword 130 ORI GI NAL WINDOW Scalepack keyword 132 INDEX PROFILE TEST Scalepack keyword 133 UNIT CELL is included in the header 139 auto editing Q amp A 142 auto dat command file 13 29 157 example 26 autoindexing 5 19 23 24 25 102 142 153 autoindexing O amp A 143 BOX and SPOT Denzo commands 29 Denzo outputs 14 best fitting lattices 144 Denzo parameters 29 don t keep re running autoindexing 81 failure 28 FIT command 32 general description 24 ice rings 39 LONGEST VECTOR Denzo keyword 60 misindexing 30 157 oscillation range 29 PEAK SEARCH FILE peaks file Denzo command 64 peaks used in 25 rhombohedral space groups autoindexing 69 twinned crystal 31 B b factor FIX B Scalepack keyword 126 B factor 87 88 90 93 95 96 97 116 121 130 133
87. ON and Y SCALE As discussed above the sign of Y SCALE depends upon the data storage convention of the instrument As was shown in the table under the Figure 5 in Detector Conventions analysis to date about 50 of the detectors have a negative values for Y SCALE The default values in Denzo are good for all automatic detectors MacScience DIP2000 series R axis Mar in their standard configurations FILM rotation depends on the spindle orientation The standard Denzo convention is to display the image such that looking along the x ray beam the spindle axis is horizontal and comes in from the right hand side Eulerian y 90 In such situation FI LM rotation Q If the spindle is at any other y value FILM rotation y 90 Eb On the Raxis the data is pre rotated by 90 degrees when written to the data file This is a convention specific to this machine Thus for Raxis images FILM rotation x 180 Raxis only B If the film or IP is rotated about the beam axis prior to scanning then the value of film rotation will have to be determined by noting the angle of rotation 49 THE AKL MANUAL Alphabetical Denzo Keyword Descriptions This is a listing of all of the Denzo keywords and their modifiers if any AIR ABSORPTION LENGTH ANGULAR OFFSET BACKGROUND radius elliptical BAD VALUE BOX CALCULATE CASSETTE flat radius rotx roty rotz CELL COUPLING CROSSFIRE x Y XY CRYSTAL rotx roty rotz DI STANCE EI GENVALUE FILTER EN
88. OSSFIRE y 0 345 CROSSFIRE xy 0 0004 CROSSFIRE x 0 023 y 0 345 xy 0 0004 53 THE HKL MANUAL CRYSTAL Modifiers format defaults example DISTANCE format default example Defines the crystal orientation for the calculation of predicted spots rotz the three angular deviations from the reference orientation specified roty by SPINDLE AXIS and VERTICAL AXIS See the diagram and further TOLX description in Denzo orientation conventions CRYSTAL rotz valuel roty value2 rotx value3_ degrees 000 CRYSTAL rotx 2 34 roty 4 566 rotz 34 560 The distance from the crystal to the beam spot on the film in mm along the X ray beam path regardless of the 20 angle DI STANCE value_ mm DI STANCE 75 0 DI STANCE 155 453 EIGENVALUE FILTER USE IT WITH GREAT CAUTION format default example END OF JOB When attempting to simultaneously refine two or more highly correlated parameters with Denzo an eigenvalue filter is triggered and an error message appears in the log file stating that the parameters being fit were too highly correlated to be fit simultaneously The eigenvalue filter value provides a way of defeating this check by lowering the value Unfortunately just because you change the value doesn t mean that the correlated parameters are going to become uncorrelated so changing the value is unlikely to help you and may lead to some horrible result EIGENVALUE FILTER value 0 001 EIGENVALUE FILTER 0 0005 S
89. R scanners user has to know instrument he or she is using Q When looking at the Denzo output I noticed that some reflections are flagged with backgr ovfl This seems to be independent from the parameters of BACKGROUND elliptical or SPOT elliptical or OVERLOAD VALUE Are there other parameters controlling this backgr ovfl The backgr ovf1 flag may be set a an overloaded pixel but also by a high percentage of pixels rejected from background calculation On MAR scanners spots outside active area corners of the image will never have valid background and will have backgr ovf1 condition set backgr ovfl column is really sum of all reflections with any kind of problem with background one of the possibilities being having one or more overflowed pixels in the background areas Other problems are background varying too much or too many pixels being obscured by other reflections or too many pixels not having a valid measurements The last affects mostly spots outside active area of the detector Adjusting RESOLUTION LIMITS can prevent prediction of spots there OVERLOAD VALUE has only effect on one of the factors involved Q When you say outlier rejection what exactly do you mean I tried changing OVERLOAD VALUE to 40000 but it doesn t make much difference Should ERROR systematic be increased then I integrated with it at 5 0 should I try with 10 0 say Neither OVERLOAD VALUE nor ERROR systematic have any direct impact on outlier rejection OVER
90. ROR SCALE FACTOR Scalepack keyword 151 estimate of I 137 Ewald sphere 33 153 experiment dat command file example 26 experimental parameters 13 experimental procedure 7 EXTEND PARTI ALS Scalepack keyword 123 F F 2 149 F and F 143 factors which lower R merge 89 Fast Fourier Transform 24 Fcalc 138 FI DUCI ALS Denzo keyword 56 FILE Scalepack keyword 124 film optical density 55 FI LM Denzo keyword 56 58 73 FILM ROTATI ON Denzo keyword 25 filter 6l final statistics 7 findbeam dat command file 42 fit algorithm 33 FIT B Scalepack keyword 126 fit cell 69 fit command first round 32 FI T Denzo keyword 57 FIT cell modifier description 57 FI T Scalepack keyword 124 FIX B Scalepack keyword 126 FIX Denzo keyword 57 FIXED WINDOW Scalepack keyword 126 Floor Down XdisplayF button 19 Floor Up XdisplayF button 19 161 The HKL Manual FORMAT Denzo keyword 58 62 63 FORMAT Scalepack keyword 127 Fourier map 8 FRAME WIDTH Scalepack keyword 127 Frame XdisplayF button 20 Fuji scanners 81 Full Scale XdisplayF button 19 G G0 Denzo keyword 58 GO statement 32 Go XdisplayF button 19 goniometer 23 80 graphite 61 Gravity beam system 48 Green Box 21 guard 63 H Hamlin SDMS gt detectors 126 127 hand of the structure 48 Heavy Atom Search 98 Help XdisplayF button 20 histogram in Denzo log file 35 history of Denzo 4 HKL MATRI X Scalepack keyword 127 HKL SCALE Scalep
91. Rmerge are provided The Bayesian reasoning and multi component error model facilitates obtaining proper error estimates b Visualization of the process plays two roles it helps the operator to confirm that the process of data reduction including the resulting statistics is correct and it allows one to evaluate problems for which there are no good statistical criteria Visualization also provides confidence that the point of diminishing returns in data collection and reduction has been reached At that point the effort should be directed to solving the structure The methods presented here have been applied to solve a large variety of problems from inorganic molecules with 5A unit cells to a rotavirus crystallized in the form with cell dimensions 700x1000x1400A Temple B Harrison S C private communication The overall quality of the method has been tested by many researchers in successful applications of the program to MAD structure determinations THE DENZO MANUAL THe D ENZO Manual A Description of the Program Denzo An Oscillation Data Processing Program for Macromolecular Crystallography 11 THE HKL MANUAL 12 DENZO FOR DUMMYS Denzo for Dummys If you have a good crystal and just want to index your frames without knowing anything follow this strategy This is not the only way to process images with Denzo nor it is necessarily the best fastest or most efficient but it covers the basics Sorry but if your proble
92. S SOFTWARE PRODUCT LICENSE I SYSMGR PLEASE SEE YOUR SYSTEM MANAGER TRACE F TRACEBACK SYMBOLIC STACK DUMP FOLLOWS Install properly your window manager Consequences of Profile fitting errors in Denzo 148 Q this line I reduced some mar 30cm data with Denzo and things seemed to go OK I am now having trouble with Scalepack It seems to reject many reflections 1 count of observations deleted manually 2 count of observations deleted due to zero sigma or profile test 3 count of non complete profiles e g overloaded observations 4 count of observations deleted due to sigma cutoff 5 count of observations deleted below low resolution limit 6 count of observations deleted above high resolution limit 7 count of partial observations 8 count of fully recorded observations used in scaling seems truncated 2 IP fitted no 2 LeOIG0 O200 0 4520 0 V 0 886 1559 IP fitted no o 3 09419 0 27 0 3581 0 0 924 1503 IP fitted no o 4 0 9741 0 30 0 3632 0 0 0 O 881 1551 IP fitted no o g O 9717 O 67 U 3626 0 0 841 1500 IP fitted no o 6 0 9535 02 79 0 3736 0 0 O0 817 1450 LP fitted no O T 0 9241 0 78 O0 3916 0 0 0 821 1364 von Stuff deleted HKLQ amp A but yet gives a fairly good r symms Shell Lower Upper Average Average Norm Linear Square limit Angstrom I tror Stats CHI 42 R fac R fac 9900 2a TT 4284627 1428 2 2860 12 222 0 3061 2105 Zedd Zac POJAL 62524 B24 122 300 Ol le
93. Scalepackvirus 9x10 unique reflections Below the maximum values of several parameters are shown for different scalepack versions Scalepack Scalepack Scalepack Scalepack Scalepack Scalepack 8m 16m manyframes virus observations 2x10 8x10 16x10 2x10 32x10 possible hk 40 000 40 000 60 000 40 000 60 000 pairs unique 1x10 2x10 3x10 1x10 9x10 reflections fitted 2x1500 21500 2x 1500 2x2000 2x2500 parameters scale amp B factor pairs batches 2000 2000 2000 4000 2000 rejected spots 50 000 50 000 100 000 50 000 200 000 Warning the bigger version the more swap memory Scalepack needs We recommend 1 5GB swap space If you have problem larger than scalepack limits the trick is to subdivide the problem into smaller batches produce an output file and use this to scale the remainder of the data Authors would like to receive e mail about inadequacy of Scalepack versions See the Appendix Recommended Hardware of the Denzo Manual 93 The HKL Manual Scenarios Scalepack allows you to analyze the quality of the diffraction from your crystals providing the statistics which result from scaling and merging the raw diffraction data Scenarios presented in this chapter are designed to show you how to get accurate data and statistics in situations typically encountered in protein crystallography These scenarios consist of input Scalepack command files followed by some comments about the statistics whi
94. TION PROBABILITY 0 0001 SPACE GROUP P21 UNIT CELL S39 84 67 S10 96 90 OUTPUT FILE mydata sca REFERENCE BATCH 101 POSTREFI NE 10 reject WRITE REJECTION FILE 0 9 FORMAT xengen urf SECTOR WIDTH 5 0 FRAME WIDTH 0 2 PROFILES SUMMED FILE FILE FILE FILE 1 OOH deg IEE LOL ctbn2 urf 201 ctbn3 urf 301 ctbn4 urf number of resolution shells for statistics ESTI MATED ERROR for each resolution shell Multiplicative factor applied to input os expected fraction of outliers in the data Here expect 1 10 000 will be outliers UNIT CELL needed because not in header OUTPUT FILE containing h k L I REFERENCE BATCH in scaling procedure Can also leave out and let eigenvalue filter set reference scale and B factor can t POSTREFI NE because not Denzo data file with list of rejected reflections from previous round of Scalepack observations with greater than 90 chance of being outliers are written to the reject file FORMAT of the input intensity data must specify these two because not in header since didn t run Denzo this may be better than PROFILES FITTED urf files are an intermediate file format created by xengen this would be data from 4 crystals or perhaps 4 sweeps of the same crystal A sweep can be a full data set Because the file format is not from Denzo you will not be able to POSTREF NE your data to get better values of the cell constants Partials are not summed in these case
95. The keyword USE PARTIALS POSITION tells the program to use a reflection that is predicted to be only partially recorded in the positional refinement A spot is predicted to be partially recorded based on the values entered for the oscillation range as well as the crystal mosaicity and beam crossfire This is a very useful feature of Denzo because it allows one to refine the orientation of the crystal even if none of the reflections are fully recorded This occurs if one collects either very thin oscillation sectors in an attempt to minimize background or has a crystal of relatively high mosaicity or some combination of both USE PARTIALS NO POSITI ON is mostly of historical interest default USE PARTIALS POSITION examples USE PARTIALS POSITION USE PARTIALS NO POSITI ON VERTICAL AXIS ORIENTATION AXIS 2 The reciprocal lattice vector perpendicular to both the spindle axis and the X ray beam Equivalent to ORIENTATION AXIS 2 format 3 integers representing the reciprocal vector default 1 0 O a axis example VERTICAL AXIS O 1 O b axis WAVELENGTH X ray wavelength Same as LAMBDA format WAVELENGTH value_ Angstroms default 1 5418 CuK example WAVELENGTH 1 5418 72 ALPHABETICAL KEYWORD DESCRIPTIONS WEAK LEVEL This is the intensity rejection level expressed as an I o ratio Only strong spots with an integrated intensity i e I above the WEAK LEVEL times the intensity error measurement 1 e are used in the posi
96. There are two possible conventions within this intuitive set In this system the X ray beam is defined as being perpendicular to the direction of the force vector due to gravity down The beam is the z axis gravity is parallel to the y axis and the x axis is perpendicular to both of these In terms of crystal orientation rotations E otz would denote rotations or the crystal around the beam axis E roty would denote rotations of the crystal about the vertical axis and E rotx would denote rotations of the crystal about the horizontal axis J Ud This convention has the advantage that it does not change with different camera spindle geometry In addition it corresponds directly to the common orthogonal detector directions in that the detector directions x and y on the film or IP are perpendicular to the x ray beam A disadvantage of this system is that the x axis may or may not correspond to the spindle axis depending upon the y chi and omega setting angles of the 3 axis goniostat For the purposes of aligning your crystal this makes life more complicated as we are used to thinking about aligning crystals by moving the arcs on a goniometer head For y equal 90 or 270 degrees and Q equal to zero however the spindle axis is parallel to the x axis and thus rotx corresponds to rotations about the spindle axis Although this system will not formally work in Outer Space where the gravity vector is essentially zero nor in the ev
97. Y SCALE gt format default example Y SCALE in Denzo serves two purposes First it is a dimensionless correction factor for anisotropy in the pixel dimensions It is applied to the fast scanning or y dimension of the pixel If the pixel is larger in y than in x then Y SCALE is smaller than 1 It is possible to create a data file which has the inverted direction of scanning along one of the axis similarly as in case of film scanned from back side Denzo handles this aberration by changing the sign of the Y SCALE keyword to negative If this is not done then the assignment of the Friedel mates Bijvoet differences will be inverted and thus the hand of the structure will be similarly inverted see Determining the values of FILM ROTATION and Y SCALE section in Denzo Orientation Conventions chapter The sign of Y SCALE is a detector property while the absolute value of Y SCALE seems stable for all detectors Once the correct sign has been determined the calibration of Y SCALE is done by refining with a high symmetry crystal trigonal tetragonal hexagonal or cubic For CCD and spiral scanners the value of Y SCALE should be exactly and not refined Y SCALE value_ dimensionless detector dependent defined by the FORMAT statement Y SCALE 1 03152 YORK OUTPUT FILE 74 Substitutes for the OUTPUT FILE command and tells the program to write the x file in the Eleanor Dodson York file format Note that S
98. YSTAL rotx is defined relative to spindle angle for which oscillation angle is 0 and it should not change when advancing to the next frame within a batch job However when I processed the data frame by frame manually the CRYSTAL rotx values changed each time at the end of the refinement no matter what value was input before the process Should the CRYSTAL rotx values be different for the different frames only different by an oscillation step saying 0 5 or should they be the same What I have seen is that the CRYSTAL rotx values are the same for each frame when the job is submitted as a batch one while they are quite different not within an oscillation step when the data is processed manually frame by frame If the CRYSTAL rotx values are different for every frame how can one include partial reflections during the process of scaling What counts is only sum of the OSCILLATION start CRYSTAL rotx It seems that you are inconsistent in how you describe crystal orientation using different procedures In the sample input of Scalepack scale in you sent me a while ago there are statements for postrefinement After the postrefinement do you suggest a reintegration It depends If mosaicity from postrefinement is higher than that set in Denzo then definitely yes If you want to precisely characterize the detector you may wish to rerefine and reintegrate the data with unit cell parameters fixed to postrefinement results Detector parameters refined
99. a detector Data are organized as unformatted files Headers if any are ignored One record corresponds to a vertical or horizontal stripe depending on the detector one pixel wide with one or two bytes per pixel Pixel values are interpreted as integers which have no sign to them The orientation conventions used by various detectors and the Denzo convention are discussed below the control data a series of keywords followed by modifying words and or some numbers and a few command words These are usually written in a command file macro file although they can be typed from the keyboard All data are read in free format Capital letters are converted to lower case unless in single quotation marks so one should be aware when inputting case sensitive UNIX file names Some keywords contain a space in their name for example Y SCALE This space must be maintained and cannot break the line Where a word or a number modifies a keyword a space comma left or right parenthesis carriage return or equal sign must separate the two These characters are recognized as separators in Denzo Comments can be put in square brackets as in this example If you wish to include a separator character in a datum for example in a title or in the name of a file the whole datum must be put within single quotes the title my crystal must be enclosed in single quotes due to the space between the words my and crystal Similar problems crop up with file n
100. ack There will be more discussion of space group determination in the Scalepack manual line pointed to by red is very important as it gives the crystal orientation parameters lines give alternative choices for these values based on crystal symmetry Thus for example if you preferred to work with small numbers you would choose the values from the second of these lines for your crystal orientation parameters It doesn t really matter it s what you prefer Line gives the X BEAM and Y BEAM positions However since the success of autoindexing is critically dependent on knowing this accurately in advance these values should be close to the initial values in the file site dat Occasionally the autoindexing routine does not give clear cut results There are four classes of data reduction problems Utter failure Misprediction Kind of failed The predicted positions agree perfectly with the peaks but data scale poorly means that there is no reasonable predicted lattice the crystal unit cell parameters are unreasonable and the predicted diffraction pattern does not correspond to the observed image In most cases failure can be ascribed to problems in three areas experimental conditions the parameters of autoindexing the peak search rarely Problems with defining experimental conditions account for the vast majority of the problems with autoindexing single crystals If any of the following parameters is widely off your a
101. ack keyword 128 HKL SHIFT Scalepack keyword 128 hklpredictions 18 22 73 reference Denzo 4 I I and I 112 BOX Denzo keyword 58 ice rings 25 29 39 58 135 142 icing 142 Ignore 153 GNORE Denzo keyword 58 I GNORE OVERLOADS Scalepack keyword 128 imaging plate 59 MAGI NG PLATE Denzo keyword 56 58 include no partials 130 INCLUDE OVERLOADS Scalepack keyword 128 include partials 130 indexing 8 manual indexing 78 infinite loop 155 INITIAL B FACTOR Scalepack keyword 122 128 INITIAL SCALE Scalepack keyword 129 NPUT Denzo keyword 59 input file format 46 INPUT Scalepack keyword 129 Int Box XdisplayF button in ZoomWindow 20 162 intensity 20 displayed in XdisplayF 21 maximum pixel intensity in XdisplayF 21 INTENSITY AVERAGING WI NDOW Denzo keyword 59 interactive processing by Denzo keywords required for 76 interactive session 41 International Tables 117 K keyword format conventions 25 keywords defaults 46 Keywords with no Defaults 76 L LAMBDA Denzo keyword 59 72 lattice superlattice 31 LATTI CE Denzo keyword 60 Launching the program 17 LEFT MARGIN Denzo keyword 60 linear and square R merge 96 Linux operating system 77 LI ST Denzo keyword 34 60 local scaling 130 log file 15 65 LONGEST VECTOR Denzo keyword 29 60 143 Lorentz polarization factor 6 lune 29 lysozyme 8 M MacScience 51 MacScience DIP gt detectors 55 MAD 130 MAD s
102. al pixel values so one can check among other things that the exposure level is appropriate If problems exist with the detector or other components of the data collection system the display option helps to discover these before all the data are recorded The examination of the image may reveal if there are extraneous sources of X ray background There are other statistics that can be provided instantly by XdisplayF which may indicate for example A D converter malfunction If there are many diffraction maxima in the image that form a characteristic pattern of diffraction from a single crystal then the next step is deducing a crystal lattice that accounts for such a pattern This step is called indexing Starting Denzo Denzo is run concurrently with the display program from another window let s call it DenzoWindow on a multi window workstation To run Denzo type something like this gt usr users students HKL denzo Make sure you are in the same directory from which you launched XdisplayF because the two programs share information through the hklpredictions and peaks file files and other X window data structures called atoms The program has no prompt so when you get a blank line the program is ready to accept your input be it commands or a macro file Since most interactive Denzo sessions start out with trying to index a particular diffraction pattern that s what I ll discuss next 22 INDEXING Getting that First Orienta
103. alculated unit cell will be quite wrong The length of unit cell along the beam direction will be 5 shorter in the perpendicular directions 10 shorter If the crystal is diagonally oriented no principal axis along the beam direction than the apparent angles between axes will violate the lattice symmetry Incorrectly defined angles between detector and the beam will result in incorrect angles between crystal axes Therefore the interpretation of the lattice symmetry is dependent to some extent on how precisely the detector parameters are known a priori Manual indexing See also Manual indexing When autoindexing fails you can try to index oscillation images manually but if you tried appendix the suggestions made above you are unlikely to do any better Manual indexing used to be the only way to run Denzo but autoindexing has superseded this difficult task Autoindexing succeeded If the autoindexing looks good then it s often nice to finish up the frame you have been working on to make sure you have the correct values for MOSAI CI TY and SPOT shape This can be done interactively in the DenzoWindow by calling a refinement macro examining the superposition of the preds on the diffraction image and then testing the MOSAI CITY and SPOT parameters to make sure all the reflections are indexed Multiple crystals simulation and interactive indexing Autoindexing can fail when more than one crystal contributes to the diffraction image So
104. aluel_ degrees end start range examples OVERLAP modifiers format default example ALPHABETICAL KEYWORD DESCRIPTIONS step range OSCILLATION range 1 0 start 30 0 OSCILLATION start 30 0 end 31 0 OSCILLATION range 1 0 step 0 8 start 30 0 the first two examples describe the same thing The third describes a series of 1 degree oscillations with ranges which overlap by 0 2 degrees Overlap defines which pixels of any spot are checked for overlap with the pixels of a neighboring reflection box The choices are the spot pixels the guard pixels found between the spot circle and the background circle or none For example if you specify OVERLAP spot and if the spot pixels overlap with the spot pixels of another reflection then the two reflections are flagged as overlaps spot guard none OVERLAP modifier none OVERLAP spot OVERLOAD VALUE format default example The lowest optical density reading considered unreliable due to saturation of the film or image plate The default value corresponds to the full dynamic range being useful with only the highest value being considered overflow A value that is higher than the maximum possible in the data tells the program to ignore the intensity overflow The range of the values themselves are determined by the analog to digital converter used to write the data For 8 bit ADC s found on film scanners the highest value is 28 1 255 For the 16 bit ADC s found on most
105. always on unless you turn it off Those of you with sharp eyes will have noticed that the auto dat macro already included the FIT crystal rotx roty rotz cell x beam y beam statement but it doesn t hurt to say them again The program uses one of two target functions specified by the keyword FIT In the first cycle following autoindexing the target function uses the reflections found by the peak search In all subsequent cycles the target function includes all reflections which overlap predicted reflections The first target function which follows autoindexing has a much larger radius of convergence at a tradeoff of less precision On all subsequent FI T cycles a more Turning FIT off sb If things get worse Finishing the Refinement REFINEMENT precise fit algorithm is used which completely ignores the list of reflections from the initial peak search Generally once the refinement is progressing you don t need to turn it off Don t set up multiple steps of fitting and fixing in an attempt to refine a single parameter at a time That is neither necessary nor smart since Denzo can fit multiple parameters simultaneously and because single parameter refinement may not allow correlated parameters to be refined sufficiently When you really don t want to refine a parameter any more for example to simulate an experiment use the command FI X as in FIX crystal rotx roty rotz GO The crystal rotation parameters will now be fixed at their
106. ames When in doubt or if the program says it can t read the data file put the file names in single quotes A different type of parser pitfall is discussed for the keywords END FIT and END FIX See the documentation under the keyword FIT Most of the KEYWORDS in Denzo require that a modifier and or a value follow For example in CRYSTAL rotx 6 012 the keyword CRYSTAL is followed by the modifier rotx which is in turn followed by the value 6 012 Floating point and integer data should follow the FORTRAN convention Text strings are limited to 80 characters When an uninterpretable input is encountered the program produces an error message and continues Omitting a necessary input causes the program to write a warning to the output file and although the program may continue it is likely to stop with an error message or maybe crash Although there are a lot of keywords in Denzo the program provides defaults for most of the critical ones that are not specific to the crystal itself These are listed in the documentation for each keyword in this Chapter and a more concise tabulation is presented in List of default parameters appendix just after the keywords list Keep in mind that just because Denzo provides a default doesn t mean that it is best to use that default The sample command files in Chapter Denzo Session and the listing in List of default parameters appendix give a listing of the keywords you should ordinarily state in your command files
107. an also leave out and let eigenvalue filter set reference scale and B partials in batches 1 through 90 summed number of cycles of postrefinement refine unit cell parameters based on all data from batches 1 through 90 space group dependent e g don t refine alpha for monoclinic space groups crystal orientation of spindle axis angle will be refined for each batch frame separately This allows you to spot slippage amp other problems one mosaicity value will be fit format of the input intensity data these values will be substituted in for the batch numbers will start at 1 for file myxtal001 x and increase to 90 for file myxtal090 x m Proper scaling requires multiple rounds of Scalepack as the statistics may improve as outliers are rejected On the second and all subsequent rounds of Scalepack the brackets would be removed from the REJ ECT command on line 9 and outliers rejected in previous rounds would be removed from the data prior to refinement of scale and B factors It usually takes about 1 to 3 iterations before no more reflections are rejected and the process converges The log file should be examined after each iteration In particular the errors y and both R factors should be checked You can adjust the error estimates in the command file to improve the accuracy of the statistics Here are some helpful guidelines Examine the difference between the total error and the statistical error in the final table
108. and click on eee to toggle between the update and remove predictions modes The display program will read the latest hklpredictions file and display the predicted reflections If the predicted reflections line up with the spots then that s one way you know that things are working Note that if you already used Peak Sear in the present session the program will automatically switch itself into the update predictions mode After you run the little autoindexing routine you get a lot of screen output The critical part is the lattice and unit cell distortion table and the crystal orientation parameters An example is shown below INDEXING Vector lengths in autoindexing from 5 7 to 152 Angstroms primitive cubic 60 08 92 79 92 79 92 79 90 00 90 00 90 00 77 48 124 78 65 26 89 82 32 01 89 54 I centered cubic 31 25 130 81 130 81 130 81 90 00 90 00 90 00 130 97 136 43 124 78 90 00 161 72 84 76 F centered cubic 37 67 146 44 146 44 146 44 90 00 90 00 90 00 146 35 146 67 146 31 63 19 127 02 147 44 primitive rhombohedral 9 82 132 55 132 55 132 55 27 55 27 55 27 55 131 62 140 63 124 78 27 65 18 16 32 85 58 37 Dead 360 56 90 00 90 00 120 00 primitive hexagonal 9 14 71 24 71 24 124 78 90 00 90 00 120 00 76 75 65 26 124 78 90 18 90 15 147 65 primitive tetragonal 11 96 54 53 54 53 124 78 90 00 90 00 90 00 41 07 65 26 124 78 90 18 90 58 389 40 I centered tetragonal 6 51 71 24 71 24 252 50 90 00 90 00 90 00 76 75 65 26 252 50 90 0
109. ant and compares the error models Basically this means that you are adjusting your estimates of the errors associated with the measurement until the deviations within observations agree with the expectation based on the error model The second mode keeps the error model constant and compares X s This mode is computationally much faster and is used in refinement procedures Of the two modes the first is more informative because it forces you to consider changes in the error model Which mode you use generally depends on what you are comparing When assessing the general quality of your detector the first mode is used When comparing a derivative to a native the second mode is used due to an incomplete error model which does not take into account important factors like non isomorphism Thus the X of scaling between a native and a derivative provides a measure of non isomorphism assuming the detector error is accurately modeled for both samples Rierge WAS historically used as an estimate of the non isomorphism of data collected on film using several different crystals and for this purpose it still has some validity because we do not account for non isomorphism in our error model It is not so important now that complete x ray data sets are collected from single frozen crystals One of the drawbacks of using Rmerge as a Measure of the quality of a data set is that it can be intentionally and unintentionally manipulated Unintentional factors
110. ape can be symmetric also BACKGROUND radius 1 5 USE BEAM only ignores all spots except the beam spot X BEAM 100 0 Y BEAM 102 0 starting values FIT x beam y beam begin fitting the beam GO GO GO BOX 2 5 2 5 narrow down box now SPOT radius 1 0 smaller spot too BACKGROUND radius 1 0 GO GO GO more refinement of beam position SPOT elliptical 0 6 0 7 0 0 normal spot size now BACKGROUND elliptical 0 7 0 8 0 0 USE NO BEAM Once the beam is found it is best not to refine x The reflections which were ignored with USE BEAM only will be refined once the RESOLUTION LIMITS are given in refine dat Another problem sometimes encountered with off line scanners involves not the scanner but rather the carousel on which the imaging plate cassettes are mounted during data collection Sometimes these old war horses carrying their heavy load of IP cassettes do not stop precisely at the zero position such that the imaging plate is perpendicular to the beam axis In this case care should be taken to refine the CASSETTE roty value Understanding the DENZo Log File 42 The log file can be examined to see if the data processing went well It is very long since it contains a record of every refinement cycle The interesting part comes after the CALCULATE statement at the end of the refinement After printing out the final refined parameters the following output is given Notes are in green 1 These numbers below are the maximum h k expected
111. as based on the mistaken idea that the variance of the pi M B p M B Ly eer L optical density value of the exposed film is independent of the degree of X ray exposure Equation 3 thus became simpler _ 2uPi M By J 2 LP 4 Many of the subsequent programs followed the Ford s formulation rather than Diamond s even when applied to data collected with proportional counters or IPs The unweighted formula proposed by Ford works quite well where the peak spot intensity 1s not much higher than the background intensity This situation arises more often with data collected on film which has a high intrinsic background and low saturation or when the crystals have low scattering power due to a very large unit cell high solvent content or a disorder The unweighted profile fitting improves the accuracy of the weak reflections compared to a straight summation but at the cost of reducing the accuracy of the strong ones This observation did lead in the past to a partial solution based on taking a weighted average between profile fitted and summed intensities where the weight is a function of the reflection intensity The weighted formula Eq 3 used in Denzo does not deteriorate the accuracy of strong low resolution reflections Thus the observed problem with the unweighted formula is in the lack of weighting Errors of the profile fitting Diamond R 1974 Profile analysis in Single Crystal Diffractometry Acta
112. at which a pixel has a 50 chance of being incorrect and thus rejected Default is 30 The background of a spot can be described by a plane and should be uniform and flat Sometimes it is not for example it may have a slope to it which can be calculated If a pixel is still out of the best fit background plane by default 30 then it is flagged as a reject It is best not to mess with this unless you really know what you are doing slope The slope of the background plane above which the entire background and spot are considered unacceptable and rejected Default is 50 units Flat 0 vertical infinite It is sometimes useful to increase this to 100 for very low background images systematic This is a way of changing the background rejection criteria when you have a major problem with diffuse scattering Default is 0 01 The range is from 0 to 1 with O meaning rely totally on the cutoff criteria and 1 meaning don t reject anything As you can see from the default you normally use the 30 cutoff criterion format REJECT fraction value defaults REJECT traction 0 75 REJECT cutoff 3 0 REJECT slope 50 REJECT systematic 0 01 example REJECT fraction 0 80 RESOLUTION LIMITS RES Defines the resolution range used in the program in Angstroms format RESOLUTION LIMITS valuelower valueupper_ A default none example RESOLUTION LIMITS 20 0 2 5 68 SECTOR format default example SKEW format default example gt SPAC
113. at is the output from Scalepack Does it contain I amp lig or F amp Fsig Why are there some negative values for I or F in the output file Output from Scalepack is I sig 1 where I refers to Lorentz polarization etc corrected diffraction intensity Sometimes this I is referred to as F 2 Even the best averaged estimate of intensity may be negative see discussion of SIGMA CUTOFF keyword in Scalepack keyword Description Other Q I cannot to input several ar files from different data collections into Scalepack Is this possible One should be able to put many different ar files I tried up to 25 Be sure that they have different batch numbers If not use the option ADD 1000 FILE XXXX1 AR ADD 2000 FILE XXXX2 AR to give different batch numbers to different ar files Q I am using an Scalepack on a VMS machine with multiwire ar files How or can I refine the mosaicity and the B factors using Scalepack You cannot refine mosaicity in ar files It has not been implemented However you can refine B factors it is done by default so you do not need to do anything special Q We run our data collection on a VMS machine and when the data is transferred to a UNIX machine the file is corrupted using ftp Some versions of ftp will create wrong file format on VMS Change to UCX native VMS implementation in such a case Also remember to use binary ftp not ASCII ftp sometimes a default option in ftp Q I sometime get a mes
114. background value was to calculate an average detector signal in the neighborhood of a specific reflection In Denzo it is assumed that the background is a linear function of the detector coordinates Robust statistics as discussed in the Numerical Recipes are applied to remove the contribution of pixels that deviate by more than 30 from the best fit to the background function If too many background pixels are flagged as outliers the whole reflection is removed from integration Denzo ignores pixels in three other cases when they have been flagged as having no measurement by an auxiliary program they have a special value e g zero in the case of R axis or MAR or they are in the spot area based on the predicted rather than the measured position of an adjacent reflection The accurate prediction of spot positions is necessary to achieve a precise integration of Bragg peaks The most important need for accurate prediction of the spot positions arises from the application of profile fitting Profile fitting is a two step process First the profile is predicted based on the profiles of the other reflections within a chosen radius The predicted profile in Denzo is an average of profiles shifted by the predicted separation between the spots so that they are put on top of each other If the predicted positions are in error then the average profile will be broadened and or displaced from the actual profile of the reflection In the secon
115. calepack keyword 132 OSCI LLATI ON Denzo keyword 62 oscillation image 25 oscillation range 8 28 29 143 low range amp still images 29 OSF 77 Otwinowski Zbyszek zbyszek mix swmed edu 4 outliers 8 135 outliers rejection 8 135 OUTPUT ANOMALOUS Scalepack keyword 132 OUTPUT FILE Denzo keyword 74 output file format Denzo _ip 149 INDEX Denzo _york1 149 OUTPUT FILE Scalepack keyword 132 Outside 44 overflows 150 OVERLAP Denzo keyword 63 overlaps 8 29 overload 55 156 OVERLOAD VALUE Denzo keyword 63 overloaded reflections 91 96 P page file quota 147 parser pitfall 46 partiality 55 partiality refinement 34 partially recorded reflections 6 81 91 partials 72 Patterson 113 Peak picking 24 PEAK SEARCH FILE peaks file Denzo command 64 Peak search XdisplayF button 19 peaks file 19 22 25 phase information 88 Photon Factory 53 Photon Factory see synchrotrons 56 plate rotation 64 polarization 114 155 POLARI ZATI ON Denzo keyword 61 POLARI ZATION Scalepack keyword 132 polymorphic crystals 144 POSTREFINE Scalepack keyword 97 132 postrefinement 28 35 Postrefinement 124 postscript file 19 potential derivative 98 precession method 6 predicted decrease 40 predictions 14 PRINT Scalepack keyword 133 print zones 144 print end print Denzo keyword 64 PRINT END PRINT Denzo keyword modifier chi square 64 fitted 64 hk1 64 memory 64 profiles 64 shifts 64 solver 64
116. calepack reads the Denzo YORK OUTPUT FILE as format denzo_ york Denzo Appendixes Keywords by Category 1 Film Pack Parameters Defined at the beginning of each job TITLE UNIT CELL LATTICE OSCILLATION MOSAI CITY WAVELENGTH LAMBDA SPINDLE AXIS VERTICAL AXIS MONOCHROMATOR RASTER 2 Film Parameters Defined at the beginning of every frame FORMAT or PL Me sek RECORD LENGTH NUMBER OF RECORDS LEFT MARGIN TOP MARGIN KEYWORDS BY CATEGORY FILM CRYSTAL CASSETTE RADI US DI STANCE Y SCALE AIR ABSORPTION LENGTH FI DUCI ALS FORMAT RAW DATA FILE SECTOR X BEAM Y BEAM 3 Refinement Parameters Defined at the beginning of every iteration step BOX SPOT BACKGROUND OMI T RESOLUTION LIMITS USE BEAM NO BEAM OVERLOAD VALUE PROFILE FITTING RADI US FIT FIX CALCULATE PRI NT ERROR WEAK LEVEL GO END OF PACK END OF JOB NPUT 75 THE HKL MANUAL Default Parameters Format independent keywords with predefined default values AIR ABSORPTION LENGTH ANGULAR OFFSET BACKGROUND CASSETTE rotx roty COUPLING CROSSFIRE x y CRYSTAL rotz roty rotx DI STANCE EIGENVALUE FILTER ERROR overload positional y x factor partiality systematic FILM rotation FIX I O BUFFER LAMBDA 860 LATTI CE P 0 0 MONOCHROMATOR graphite lowest reading in spot NO BEAM area OVERLAP none Tau PLATE ROTATI ON 0 0 Gwo PRINT chi square A PRINT no profiles Oe 20s a0 PRINT no solver ae PRINT shifts ee PRI
117. case of large oscillations the lunes may have severe overlap and the autoindexing program will have a problem with the non uniqueness of the mapping from detector to reciprocal space In the case of small oscillations or stills collected on non mosaic crystals with a very parallel beam too few reflections may be present to describe a lune As a rule of thumb the following types of oscillation ranges should be used viruses 0 25 oscillations proteins 1 0 oscillations small molecules 2 5 oscillations assuming well diffracting samples There is a fairly wide margin on either side at least a factor of 2 Before autoindexing Denzo edits the peak list produced by the peak search algorithm of the XdisplayF program Problems with the Denzo autoindexing parameters account for a few of the remaining problems with autoindexing These may have to be adjusted in the case of non single crystals very large or small unit cells lots of noise peaks or ice rings These parameters are controlled by the keywords LONGEST VECTOR RESOLUTION LIMITS WEAK LEVEL BOX and SPOT may need to be adjusted if the peaks are sparse due to a small unit cell or low mosaicity In this case it should be set to equal the longest vector of the reduced primitive real space unit cell If you did not understand that last sentence simply use the longest unit cell axis which is the same in primitive space groups anyway Better to err on the side of larger rather tha
118. ce The reason we focus our attention on the crystal orientation is simply because the detector or site parameters like DI STANCE 20 and the position of the direct beam at the detector are and should be reasonably well known before processing begins Thus they can be refined automatically by the program The HKL package offers two indexing methods automatic and interactive The automatic method applicable in most cases is fast and simple The first step in the automatic method is the peak search which chooses the spots to be used by the autoindexing subroutine Ideally the peaks should come from a diffraction by a single crystal The Denzo program accepts peaks for autoindexing only from a single oscillation image It is important that the oscillation range be small enough it even can be zero 1 e a still so that the lunes rings of spots from one reciprocal plane are resolved The requirement of lune separation is distinct from the requirement of spot separation If the lunes overlap spots may have more than one index consistent with a particular position on the detector On the other hand the oscillation range should be large enough to have a sufficient number of spots for the program to be able to establish the periodicity of the diffraction pattern This may require at least 0 5 degree oscillation for a small unit cell protein crystal and 2 3 degree oscillation in the case of organic small molecule crystals The second step in th
119. ce of the scaling In the default procedure if scaling does not converge in 20 default cycles of refinement B factors will be not be fitted The FIT B command can override this Not to be confused with the postrefine FIT B command described above You cannot postrefine B temperature factors FIT B this flag is turned off FIT B This flag tells the program not to fit B factors at all Usually it is combined with the input of the B factors you want to apply but do not wish to refine anymore or it is used for frozen crystals where you do not expect significant decay FIX B turned off B factors are fit after the scale factor refinement converges FIX B For Hamlin archive files a fixed window of 3 5 7 or 9 frames or the original Hamlin determined window of frames may be used for summing a reflection Valid for archive file format only format FIXED WINDOW value 126 THE SCALEPACK KEYWORD DESCRIPTIONS default uses the Hamlin definition of the window example FIXED WINDOW 7 FORMAT This keyword specifies the format of the input Akl and intensity data Input data can come from any of nine types of files This program requires this keyword to properly read the files modifiers denzo_ip from frames processed with Denzo denzo_york1l output created with Denzo option york scalepack from Scalepack output file rigaku raxis binary R Axis software output dS prorie nk binary output from XDS output file madnes procor intout binar
120. ch correspond to many of the typical applications of Scalepack If you do not find your particular problem addressed by these scenarios we hope that they along with the keyword documentation will provide enough guidance so that you can construct your own 87 THE HKL MANUAL Program Description The scaling algorithm used in Scalepack is described in Fox G C Holmes K C 1966 An Alternative Method of Solving the Layer Scaling Equation of Hamilton Rollet amp Sparks Acta Cryst 20 886 89 1 88 The scaling and merging of different data sets as well as the global refinement of crystal parameters postrefinement is performed by Scalepack Unlike in other procedures in Scalepack the estimated error of the measurement is enlarged by a fraction of the expected rather than the observed intensity This algorithm reduces the bias towards reflections with an integrated intensity below the average Due to the correlation between crystal and detector parameters the values of the unit cell parameters refined from a single image may be quite imprecise This lack of precision is of little significance to the process of integration as long as the predicted positions are on target There is no contradiction here because at some crystal detector orientation the positions of the reflections may only weakly depend on the value of a particular crystal parameter At the end of the data reduction process one would wish to get precise unit cel
121. ch result Scalepack is run in batch mode unlike Denzo which has a graphical display component and is best run interactively All of the scenarios are thus given as command files For clarity the first two or three lines of each command file which set the directory and call the program have been omitted but are given here setenv cnverr yes only for IBM RS 6000 obligatory cd frames mydir Denzo set directory to file which contains x files Scalepack lt lt STOPINPUT gt gt scale log start the program name the log file STOPINPUT It is necessary to run the program this way rather than making each of the scenarios an input file called by the command e g scalesimple inp because the Scalepack parser does not recognize nested commands This is important for Scalepack because the list of rejected reflections is usually called with the re ject command Just one of those things List of Scenarios 94 ee ee Scale data from one crystal processed with Denzo Scale data from one crystal not processed with Denzo Heavy atom search Scale native a few frames of a potential derivative Scale multiple native data sets together Data from Denzo Reindexing a Switch a and cin P2 or P2 Change the order of a b c to b c a in orthorhombic Convert a b c to c a b in orthorhombic Reindex P23 switching a b P321 or P312 reindex data and combine with another data set P6 P4 or H3 reindex data and combine with another data set R
122. ckgrounds due to saturation or excessive background editing and others represent reflections which sit on the edge of the image and are lost but not outside the detector Outside is a calculation of how many reflections are missing because they landed outside of the scanned image area i e off the IP for the specified resolution range Remeasured is the same as total on the last cycle 43 THE H KL MANUAL type good intns ovil fiducials O 0 O partials 2004 60 1556 whole Waa BGS 52 740 total A743 112 2296 O O O O 5 These are the final statistics for the positional refinement spots refined 4855 chi 2 refl partiality refinement Understanding the DENZo x file x 0 70 y 3620 chi 2 refl 0 78 pred decrease 0 80 pred decrease weak overlap backgr ovfl others total outside remeasured ac O 1 O 1 1 4 3625 850 2020 O 0 2931 Zio Faad 2 d FLN 1123 TLJ 0 000 4855 II O O 0 000 3620 0 0 The product of Denzo is a file called the x file named for the suffix it is usually given Each file represents the distillation of a single image The x file is organized into three consecutive parts header listing of reflections and refined parameters Here is an annotated example ajit alpha toxin a matrix representation of orientation title of the frame u matrix orientation 0 00265575 0 00167628 0 00863879 0 069715 0 985799 0 152772 0 00877937 0 00028329 0 00281712 0 289696
123. cle x5 mm y5_ mm mm triangle x6 y6 x7 y7 x8 y8 all values in mm nothing ignored IGNORE quadrilateral 50 0 52 0 52 100 50 100 IMAGING PLATE REDUNDANT INPUT format default examples This substitutes 1 for 1 with the keyword FI LM See FI LM for all the details This keyword is the same as the keyword Signals to the program to change reading of the control data to an auxiliary file At the beginning of the program the auxiliary file is closed so the keyword NPUT has to be followed by the filename of the auxiliary file The auxiliary file is read until a semicolon is reached or an end of file mark is detected The control is then returned to the original input stream In the case of reaching a semicolon the auxiliary file remains open and the next time the keyword NPUT appears it must not be followed by a file name or else the program will become confused since it still wants to read from the open auxiliary file On the other hand if control is returned due to an end of file mark then the next time input is encountered it must have an auxiliary file name next to it If you give it the same auxiliary file name as before then it will begin reading the auxiliary file from the beginning Nesting of auxiliary files is not allowed Auxiliary files are useful to keep the data common to many films in a separate file INPUT filename none INPUT subbatchl dat refine INTENSITY AVERAGING WINDOW Sets the number
124. cluding consistency indexes e g Rmerge but not necessarily the correctness of the merged data Detector geometry X ray beam See Arndt U W X ray Collimation and Genera tion in Data Collection and Processing Daresbury 93 DATA COLLECTION PROCESS If the detector marginally passes a standardized test it may fail during data collection For example if the test data are collected in the large oscillation angle mode a shutter opening delay or spindle motor backlash may affect fewer partials than if the data are collected in the narrow frame mode It is dangerous to accept results from the test with a significant number of reflections flagged as outliers even if the Rmerge Statistics seem to be good This is almost a sure sign of a problem and unless the problem is well understood it may not be sporadic when one collects data to solve a crystal structure One way to attempt to understand the nature of outliers is to locate them in the detector space in order to identify the problem The clustering of outliers in one area of the detector may indicate a damaged surface if most outliers are partials it may indicate a problem with spindle backlash or shutter control The Zoom mode may be used to display the area around the outliers in order to identify a source of problem for example the existence of a satellite crystal or single pixel spikes due to electronic failure Sometimes a histogram of the pixel intensities may suddenly s
125. control input Individual measurements can be rejected explicitly in the control input Control input is like that for Denzo each record has a title and the order matters only to some degree The title tells what input is expected to follow All file names must be 80 characters or less and it is usually safer to enclose all file names in single quotes myxtal x Usually input variables follow in the same record as the title except where the input is a table or a list Anything between and in the command line is considered a comment by the program A semicolon terminates input Unlike Denzo Scalepack is usually run as a batch job because there is no point in watching your frames scale and because the Scalepack log file is where the statistics associated with scaling and merging are presented Statistics and Scalepack The quality of X ray data is initially assessed by statistics In small molecule crystallography there is almost always a large excess of strong data so this allows the crystallographer to discard a substantial amount of suspect data and still accurately determine a structure Compared to small molecules however proteins diffract poorly Moreover important phase information comes from weak differences and we must be sure these differences do not arise from the noise caused by the limitations of the X ray exposure and detection apparatus As a result we cannot simply throw away or statistically down weight marginal da
126. ct of the first and second integers PRINT PRINT PRINT PRINT no profiles shifts no shitts solver no solver statistics Jno statistics defaults chi squares printed after every cycle of refinement no fitted spots printed no hk1s printed print memory statistics on the last cycle no profiles printed except on the last cycle when a 3 x 3 grid is printed all shifts printed no solver statistics turned on on the last cycle only examples PRINT PRINT fitted 12 6 U profiles 3 3 tells the program to divide the frame into 9 equal regions and print out 9 average spot profiles the more profiles you specify the larger the log file will be PRINT hkl 12 8 0 You can easily generate gigabytes of log file if you use some of the print options so be careful 65 THE HKL MANUAL PROFILE FITTING RADIUS The radius of the area around each particular spot in mm containing neighboring spots used to calculate the average spot profile The spot in question is fitted to the average profile of all the spots within the specified radius Generally the radius is set so that spots on roughly 3 5 of the area of the detector are included in the averaging Only on the last cycle of the refinement is the profile fitting applied via the keyword CALCULATE The particular size of the area chosen depends on how much the profiles vary across the film and the density of the spots on the film For example if the spot profi
127. cted partiality multiplied by an average intensity in the same resolution range 34 REFINEMENT The formula for the residual difference between expected and predicted value is the same as in the postrefinement however at this stage the error of the predicted fully recorded intensity is much larger equal to the expected intensity Nonetheless the concomitant positional and partiality refinement used in Denzo is both stable and very accurate The power of this method is in proper weighting by estimated error of two very different terms one describing positional differences and the other describing intensity differences The method leads to a reduced correlation between the detector and crystal parameters An additional benefit is the uniform treatment of both detector and crystal variables in the whole refinement process Mosaicity and Spot Shape Spot Shape and Size Denzo defines mosaicity as the smallest angle through which the crystal can rotate about any axis or combination of axes while a reflection is still observed Typical values range from less than 0 1 degree to several degrees Generally the higher the mosaicity the less useful the crystal Denzo does not refine mosaicity so you must input a reasonable value The easiest way to do this is to simply increase the mosaicity entering it as MOSAICITY 0 5 GO for example until all of the observed reflections are matched with preds Thus if you have more observed reflections
128. ction of outliers in the data reflections with gt 90 probability as outlier are written to the rejection file may use the first frame as reference for scaling or can omit this and use eigenvalue filter reject outliers from previous rounds 10 cycles of postrefinement there are missing batch numbers created by the file Statement below but we do not need to mention this for postrefinement note which parameters are fit over the two data sets and which are for each frame the consecutive sequences must be declared here otherwise program will try to ADD PARTIALS from non consecutive frames first data set has 103 frames second data set sectors in batches starting with 201 SCENARIOS Alternatively if you have already processed the two data sets and have two sca files you want to scale and merge them you can use the following command file Eb Note that you cannot POSTREFINE or ADD PARTIALS between the two data sets You are basically scaling and merging one set of processed and outlier rejected native data to another scaletwosca com NUMBER OF ZONES 10 ESTIMATED ERROR 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ERROR SCALE FACTOR 1 0 FORMAT Scalepack Data Set i x FILE 1 my_crystall sca Data Set 2 x FILE 2 my_crystal2 sca OUTPUT FILE my CrystellZ sca number of resolution shells for statistics ESTIMATED ERROR for each resolution shell Se
129. d applied to I to give Icorr Tor is a solution to the following equations D sb _ o S E Leo Eo where E4 and E2 are the input variables error scale factor and estimated error respectively and PROGRAM DESCRIPTION Legri W og a The output o I is l r During parameter refinement the scale and B if requested for all refinable films are shifted simultaneously to minimize the difference between the OER a and Tceorr s for individual measurements summed over all reflections on all films 7 s are redetermined in each cycle as described above One or more of the batches that is input is designated as the reference batch and its scale and B factors are not refined There is room for many reference batches this program will also work with no reference batch in which case eigenvalue filtering stabilizes the shifts in scale and B factors Initial default scale and B factors are applied to each film this can be replaced by explicit input of initial values Initial scale factors should be around 1 0 or larger for very strong data so the output values are not too large otherwise they can overflow fixed formats The user controls whether anomalous pairs are treated as equivalents or not and whether they are merged The program rejects measurements that fail an I o 1 test when they are read in and these do not contribute to the output Partially recorded reflections can be added or not according to the
130. d dd conv swab Better to do it in Denzo Program dd may incorrectly change header information that may be useful On SGI one must load the files using dev tapens which tells it not to swap the bytes dev tape the default does a byte swap to conform to ANSI standards To wit if you try to load using dev tapens tar will complain about checksums and formats and then abort Load using dev tapens then use dd conv swab We would like to process our data on a VAX VMS old style machine although we can also use HP and SGI workstation Therefore we would be mostly interested in VMS version of Denzo for R axis and also synchrotron data Why use VAX if you have SGI and HP I ran the programs on a SGI running the 5 1 1 2 operating system and I see they were compiled on 4 0 5 UN csh ERROR cannot execute binary file UN csh ERROR Scalepack Exec format error The new version of Scalepack doesn t seem to run on our SGI machines I get an error message Scalepack Exec format error Wrong Architecture I ve tried IRIX versions 4 0 5 and 5 1 1 2 Any ideas You probably try to run Scalepack compiled on OSF or other brand of computer Do you have a Alpha VMS version of the latest stuff Yes however future versions 2 0 and higher will not run on Alpha VMS Error message from my computer MARD TGABND 001 IMAGE SLICENSE F NOAUTH DEC DW MOTIF USE IS NOT AUTHORIZED ON THIS NODE LICENSE F NOLICENSE NO LICENSE IS ACTIVE FOR THI
131. d step the information from the actual and the predicted profile is combined by the following process The observed profile M is a sum of the Bragg peak and background P the estimate of M is expressed by the formula P B Const p 1 where B is the predicted value of the background and pj is the predicted profile Profile fitting minimizes function h i V 2 where V is a variance o of M V is a function of the expected signal in a pixel which in the case of a counting detector is P The index 7 represents all pixels in a two dimensional profile the same formulation of profile fitting applies to one and three dimensional profiles The predicted profile can be normalized arbitrarily the most natural definition of normalization is that the sum of p is equal 1 Such a choice makes the constant in expression 1 the fitted intensity 7 1 e Z Const The solution to the profile fitting can be expressed by an alternative but mathematically equivalent approach presented below Each pixel provides an estimate of the spot intensity J equal M B p with variance V p A profile fitted intensity is then simply a weighted average of all observations 37 THE AKL MANUAL Pi Vi Pi Pi Ly Ly 5 This approach without an explicit solution presented in Eq 1 was first published by Diamond in 1969 for the one dimensional case However in 1974 Ford proposed a simplified formula where V is constant This w
132. detector raw data binned dezingering on ADSC Quantum4 CCD detector corrected data ADSC Quantum4 CCD detector corrected data binned ADSC Quantum210 CCD detector corrected data ADSC Quantum210 CCD detector corrected data binned ADSC Quantum315 CCD detector corrected data ADSC Quantum315 CCD detector corrected data binned CCD ADSC detector single CHESS red EMBL scanner in Hamburg INST HAMBURG MAR scanner LBL LMB prototype 1187 pixel format MAR scanner in MRC LMB compressed ock mode MAR scanner in small plate mode INST _HAMBURGN MAR scanner in large plate mode INST _HAMBURGD MAR 345mm scanner 180mm scan 100 micron size MAR 345mm scanner 180mm scan 100 micron size compressed MAR 345mm scanner 180mm scan 150 micron size MAR 345mm scanner 180mm scan 150 micron size compressed MAR 345mm scanner 240mm scan 100 micron size MAR 345mm scanner 240mm scan 100 micron size compressed MAR 345mm scanner 240mm scan 150 micron size MAR 345mm scanner 240mm scan 150 micron size compressed MAR 345mm scanner 300mm scan 100 micron size MAR 345mm scanner 300mm scan 100 micron size compressed MAR 345mm scanner 300mm scan 150 micron size MAR 345mm scanner 300mm scan 150 micron size compressed 84 Ver 1 96 and higher ccd adsc quantum4 binned ccd adsc quantum4 binned dz ccd adsc unsupported q4 ccd adsc binned unsupported q4 ccd adsc unsupported q210 ccd adsc unsupported q210 binned
133. dimensional diffraction images each of them representing a different curved slice of reciprocal space In order to integrate the diffraction maxima accurately they must appear as separated non overlapping spots in the individual 2 d images Unless the data are collected by the precession method the diffracted image contains a distorted view of reciprocal space This distortion of the image is a function of the data collection method the diffraction geometry and the characteristics of the detector For the data reduction to be successful the distortion of reciprocal space as viewed by the detector has to be accounted for correctly by the program The distortion of the image of reciprocal space can vary even between images collected on the same detector This is because the position of the detector the X ray wavelength the oscillation range pixel size scanner gain and the exposure level all affect the detector representation of diffraction space One should start data collection and reduction with a careful inspection of the data in their raw original form The zoom option of the program XdisplayF allows one to examine reflections in pixel by pixel detail to check that the diffraction maxima are resolved Because the program displays the resolution in Angstroms corresponding to the position of the mouse driven cursor the diffraction limit of the crystals can be estimated even without data reduction The display in high zoom mode provides digit
134. dir DENzo machine specific calls usr local bin DENzo lt lt STOPINPUT gt gt xtal log start the program name log file site dat myexperiment1 dat note modified expt macro here refine dat 4 THE H KL MANUAL findbeam dat END OF PACK end marker increment sector END OF JOB STOPINPUT unix stop signal Note that the file myexperiment1 dat has been modified from myexperiment dat to include the final refined values of everything determined in the interactive refinement of the first frame done above The easiest way to do this is to examine the xtal001 x file produced by the CALCULATE GO statement At the end of this file are all the final parameters determined from refinement Cut these out and just paste them into the end of myexperiment dat and rename as myexperiment1 dat Since you are not running autoindexing in the batch mode you have to take the results and paste them in by hand Warning for unusual instruments e g R Axis with non vertical spindle synchrotrons make sure that you have the correct value of FILM ROTATION The defaults in the formats are only for the standard configuration of each instrument All that remains is to submit the job and examine the log file for the results The processed frames in this case have the name xtalo x The wild cards have been replaced by the SECTOR argument These x files are ready to be scaled in Scalepack BOX 5 5 5 5 big box SPOT elliptical 2 0 10 0 0 0 large spot sh
135. e HKL SCALE Divides each h k and by the input values Useful for reducing the unit cell volume particularly after Akl matrix transformation Rarely needed Example index data originally in C222 and want you data in say P3 you would apply HKL SCALE of22 1 and HKL MATRIX of 1101 1000 1 so that the new indices would have values of h k 2 h k 2 and ay format HKL SCALE valuel value2 value3 default 11 1 example HKL SCALE 2 2 1 HKL SHIFT Used to make a quick test of misindexing Adds the specified integer vector to each original h k and l If you are successful in using this congratulations you now have to reprocess your data in Denzo with the correct indexing format HKL SHIFT integer integer integer default 0 0 0 example HKL SHIFT O 1 0 we ll assume you weren t too far off IGNORE OVERLOADS Opposite of NCLUDE OVERLOADS This is not the default Useful if you collect data at low and high exposures and is useful to ignore the saturated reflections at the high exposures Applies to data read in after this command INCLUDE OVERLOADS NCLUDE OVERLOADS is flag for whether fitted profiles with some pixels missing typically due to overload should be included in the scaling Affects only Denzo image plate output files formats denzo_ip denzo_york1 Note that for summed profiles this does not apply because only profile fitting can estimate the value of the overloaded pixels format NCLUDE OVERLOADS default NCLUDE OVERLOADS
136. e format ORIENTATION AXIS 2 integer integer integer default 001 example ORIENTATION AXIS 2 1 0 0 131 The HKL Manual ORIGINAL WINDOW Refers to the window size of ARCHIVE format files Opposite of FI XED WI NDOW OUTPUT ANOMALOUS OUTPUT FILE format default example POLARIZATION alias for the keyword ANOMALOUS Name of file for output of scaled measurements A new file will be created if none exists but a pre existing file will be overwritten Maximum of 80 characters allowed in the file name Depending on whether ANOMALOUS flag is set there are one or two sets of I and o I per reflection Output is A k I o 1 NO ANOMALOUS flag or h k L I o I I o l ANOMALOUS flag in format 314 4F8 1 If the NO MERGE flag has been set unmerged data are output as A k l asymmetric unit I o 1 in format 414 2F8 1 and the ANOMALOUS flag has no effect For NO MERGE original index the format is original Akl unique hkl batch film number a flag 0 centric 1 I 2 I another flag 0 Akl reflecting above the spindle 1 Akl reflecting below the spindle the asymmetric unit of the reflection I scaled Lorentz and Polarization corrected and the o 1 The format is 614 16 212 13 2f8 1 If I I or I are larger than 10 they are output in format I8 When you read the Scalepack output by a third party program check if the program uses standard fortran input f8 0 instead of non standard
137. e autoindexing is the mapping of the diffraction maxima identified by the peak search onto reciprocal space Because the precise angles at which reflections diffract are a priori unknown for oscillation data the center of the oscillation range is used as the best estimate of the angle at which the diffraction occurs 23 THE H KL MANUAL 24 Autoindexing Peak picking The autoindexing in Denzo is based on a novel algorithm a complete search of all possible indices of all reflections that are found by a peak search or manual selection When the program finds values integer numbers of one index for example h for all reflections this is equivalent to having found one real space direction of the crystal axis in this case a For this reason such indexing is called real space indexing Finding one real space vector is logically equivalent to finding the periodicity of the reciprocal lattice in the direction of this vector The search for real space vectors is performed by a Fast Fourier Transform FFT and takes advantage of the fact that finding all values of one index e g h for all reflections is independent of finding all values of another index e g k The Denzo implementation of this method is not dependent on prior knowledge of the crystal unit cell however for efficiency reasons the search is restricted to a reasonable range of unit cell lengths obtained by default from the requirement of spot separation After the search fo
138. e high resolution data is generally weaker than the low resolution data and as a result is more likely to result in higher Rmeree By down weighting the high resolution data the Ringe 18 artificially lowered Any program which does not allow the option of setting the reference frame will have this problem Of course there is no problem with non decaying crystals There are also intentional ways of lowering your Rmerge Like those ways listed above they generally result from the statistically invalid elimination of weak reflections reduction of the redundancy of the data or de emphasis of weak data The difference between these methods and those listed above is that they are generally under the control of the user Use of an unreasonable SI GMA CUTOFF e g gt 0 The rejection of weak data will always improve Rmerge There is a further discussion of SI GMA CUTOFF in the Scalepack keywords documentation section Use of a resolution limit cutoff Again the omission of weak data will improve Rmerge A reasonable resolution cutoff is the zone where I o lt 2 Combining into a single zone for the purposes of calculations those resolution shells where Rmerge 1S rapidly changing In this case the shell will be dominated by the strong data at the low resolution end of the zone and give the impression that the high resolution limit of the zone has better statistics than it really does For example if you combined all your data into a 9 T
139. e not used in refinement and as reference profiles Is WEAK LEVEL the explanation and should I rerun Denzo with an explicit value for WEAK LEVEL or would you know any other reason for my missing data below 50 Your problem comes from PROFILE FITTING RADIUS not being set You have to rerun Denzo WEAK LEVEL in Denzo has nothing to do with Scalepack keyword SI GMA CUTOFF We would like to have some information about the output files x of Denzo Regular film output file Denzo _ip imaging plate output format i e the format you would you to read in in FORTRAN is format 314 12 2f8 0 f7 0 f6 0 f6 0 2f7 0 f6 0 f8 0 hkl Flag O full 1 partial Intensity F 2 by profile fitting o of intensity F 2 A of profile fitting Intensity F 2 by profile summation Cosine of incidence angle at detector Predicted pixel position of spot centroid slow fast directions Lorentz polarization obliquity combined factor Strength of averaged profile in arbitrary units Alternative york output file Denzo _york1 format is format 314 12 2f8 0 f7 0 f6 0 f6 0 2f7 0 f6 0 f8 0 149 The HKL Manual 150 hkl Flag O full 1 partial Intensity F 2 by profile fitting o of intensity F 2 42 of profile fitting Intensity F 2 by profile summation Cosine of incidence angle at detector Predicted pixel position of spot centroid slow fast directions Lorentz polarization obliguity combined factor Strength of averaged profile in a
140. e of the proper treatment one can do approximations One of them is provided by French and also by French and Wilson and implemented in the CCP4 program TRUNCATE A very simplified and somewhat imprecise implementation of TRUNCATE 1s this if I gt o 1 F sqrt if I lt o 1 F sqrt o 1 format SI GMA CUTOFF value default example SIGMA CUTOFF 2 5 SPACE GROUP 138 Space group symbol from the list below This input is required The space group may be entered as a name e g P212121 or as a number e g 19 for the same space group Most of the numbers correspond to those of the International Tables The numbers above 230 are non standard definitions of space groups 1 Pal one P422 154 P3221 207 P432 2 PZ 299 P4212 T33 R32 208 R4227 4 PaL 91 P422 168 PG 209 F432 5 CZ 92 Pali 169 Pol 210 F4132 16 P222 93 PAZAZ TRU P65 ALI LA Oe Ly P2221 94 PaZZ12 Hey ail POZ ZA P4332 18 PZLA oD P4322 Ia P64 213 P4132 19 PZLA T21 96 PASATZ Ika P63 214 4132 20 C2221 FI T422 UROS FGAZ2 303 PAC 21 C222 98 TALZ Z ETS POL2Z2Z 305 B2 22 E222 143 PS 179 POgZ2 S16 PAL aa Ao TZAZ 144 Pel 180 P6222 401 CA 24 I2 12121 145 P22 Led P6422 403 PZG TS P4 146 RS LOZ P6322 446 H3 76 P41 149 PS LZ Lo P23 455 HoA a7 P42 130 B321 L96 F23 501 Ei THE SCALEPACK KEYWORD DESCRIPTIONS 78 P43 Lal P3L12 197 cao 503 LA ao I4 LZ Poolal Loe P213 505 CAL 80 T41 ibe a P3212 199 Tals Notes to particular space groups 146 R23 R3 in hexagonal setting 446 H3
141. ection START REFINEMENT denotes the start of the series of commands given to index the diffraction image The end of the START REFINEMENT loop is denoted by the CALCULATE GO statement The START REFINEMENT statement must come after the SECTOR x to y statement and the loop is run for every sector value START REFI NEMENT 1 format default example SPOT modifiers formats SPOT SPOT defaults none examples SPOT SPOT START REFINEMENT format default this is not a default example B sce the example command file in Finishing Up section 70 ALPHABETICAL KEYWORD DESCRIPTIONS STOP REDUNDANT This command stops the Denzo job Same as QUI T or a semicolon SWAP BYTES OBSOLETE TITLE format default example TOP MARGIN Format default example UNIT CELL format default example USE BEAM modifier All HKL programs now automatically recognize swapped data Data set title string If it contains any spaces commas equal signs or brackets DENZO recognized separators it must be enclosed in single quotes none TITLE my crystal The top margin in mm from the top edge of useful data i e the beginning of every record Do not set the top margin to be larger than the film length Note that commands like top margin are working in the data coordinate system It is very important to check the results of these commands in the display window TOP MARGIN value_ mm 0 0 TOP MARGIN
142. ects autoindexing procedure in two ways 1 the size affect the default value of LONGEST VECTOR if the LONGEST VECTOR command is absent from auto dat file 2 The spot size also affects the signal to noise ratio Only spots above WEAK LEVEL default value 5 are used in autoindexing You can see these spots in XdisplayF window If the spot size is very wrong you may not get enough spots for autoindexing After processing the first image and getting refined values for the various parameters I then entered the refined parameters in particular the values for CRYSTAL rotx roty and rotz into another auto dat file and processed a different image However the values for CRYSTAL rotx roty and rotz that come out of the indexing run are quite different though clearly related Is this normal and why does it happen I have to think a bit about whether it changes the indexing in particular whether the assignment of F and F is switched Certainly if after running the auto dat in Denzo I then redefine again the crystal angles to be consistent with those refined from the first image then the predictions look fine and the refinement proceeds nicely though in one case CrysX is 366 26 instead of 6 26 but that should not matter Any advice on this would be welcome It is normal for autoindexing with different images to result in indexings related by a lattice symmetry operation Such a rotation never changes the assignment of F and F No rotat
143. ediction of small errors Assuming default values of the error estimates y values of 2 or even 3 are acceptable because the position of the predicted reflection and hence the intensity is still very accurate Thus the y values may not reflect errors in the integration of the reflections Finishing Up refine dat REFINEMENT It is a good idea to keep your error model roughly the same from crystal to crystal so that you can use the s to pick up any other variations in your data collection For example if you are getting y s in the range of 2 to 5 consistently and your crystal is good consider problems with the detector alignment If the x values are very high gt 10 or so for default values of the error parameters something is seriously wrong with the indexing refinement or the detector This will be apparent from visual inspection of the displayed image in any case Since the goal of Denzo is to produce a list of Hk s and unscaled intensities it does not matter that the errors in the positions differ a bit from the standard expected error On the other hand x values substantially different from 1 in Scalepack should be investigated because this directly affects a critical result of the scaling procedure namely the o value assigned to each scaled intensity So that s where you should pay attention to finer deviations of y values from 1 not in Denzo Once your first frame is indexed and you ve adjusted the mosaicity
144. efinement sometimes called postrefinement can refine crystal parameters more precisely The unit cell used in further calculations should come from the global refinement in Scalepack and not from Denzo refinement Partiality refinement The refinement of the fraction partiality works by minimizing the function gt a Cn a Ian partials where G F F I m error positional I i the sigma of the measurement of the reflection Lai the intensity of the partially recorded reflection F the fraction partiality Iii the average intensity of the fully recorded reflections in this particular resolution shell error positional the error in the positional measurement see ERROR positional keyword The quantity f as a function of the crystal parameters is the variable in the equation The other parts are determined by the observations and by the error model The detector and crystal parameters are refined by a least squares method that minimizes the deviation of the reflection centroids from their predicted positions Such refinement by itself is seriously deficient when applied to a single oscillation image since one crystal rotation parameter rotation around spindle axis is undefined and the others are highly correlated and or poorly defined To overcome this problem another term partiality refinement is added within Denzo in which the intensity of the partially recorded reflections is compared to the predi
145. eindex from H32 to R32 Reindex two H3 data sets each with inconsistent index and combine sm ho ao go Reindex four P3 data sets each with inconsistent index and combine pani 0 Comparing two data sets from Scalepack Compare I and I detection of an anomalous signal Fitting polarization High resolution low resolution data sets from one crystal Thin frames or few frames SCALE RESTRAIN andB RESTRAIN Space group determination SCENARIOS Scenario 1 Scale Data from One Crystal Processed with Denzo scalesimple com NUMBER OF ZONES 10 ESTIMATED ERROR 0 04 0 04 0 04 0 04 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 3 REJECTION PROBABILITY 0 0001 reject WRITE REJECTION FILE 0 9 SPACE GROUP P21 OUTPUT FILE mydata sca REFERENCE BATCH 1 ADD PARTIALS 1 to 90 POSTREFINE 10 FIT crystal a 1 to 90 FLT crystal b 1 to 90 RIL Cisysieeull vee 1 eo S10 FIT crystal beta 1 to 90 FIT batch rocx 1 to 90 FIT batch roty 1 to 90 FIT crystal mosaicity 1 to 90 FORMAT denzo_ip SECTOR 1 to 90 FILE 1 frames denzo myxtal x number of resolution shells for statistics estimated error for each resolution shell Multiplicative factor applied to input o expected fraction of outliers in the data read in rejected reflections observations with greater than 90 chance of being outliers are written to the reject file output file containing A k L I oj reference batch in scaling procedure C
146. eir original form as well as to observe the progress of data reduction Displaying the raw data takes advantage of the human ability to grasp the significance of complex patterns that would otherwise be hard to analyze numerically Visualization of the data collection and reduction process allows efficient recognition of what is correct or wrong by taking advantage of the brain s ability to recognize complex visual patterns Problems like detector vibration may produce very striking patterns on the diffraction image yet the deleterious effect on the final Rmerge Statistics may not be so obvious The optimal experiment has to take into account the interaction between all of the following the experimental point spread function the unit cell length the detector position the angular range of data collection the mosaicity and the definition of the integration box The ability to visualize the influence of different data acquisition parameters allows the experimenter to understand his particular experimental limitations and help to overcome them Starting XdisplayF The display program for Denzo is based on the X window interface XdisplayF runs concurrently with a Denzo session either in a separate window or on a separate graphics terminal Both Denzo and XdisplayF store the entire image in memory so the host computer must have approximately depending on the exact size of the image 64 to 512 MB of RAM in excess of that needed for the operating sy
147. ell parameters and below are the undistorted unit cell parameters for comparison in case of a primitive rhombohedral lattice there are actually two lines the second with parameters in hexagonal indexing As you can see for some lattices e g primitive tetragonal all the cubics etc a substantial amount of distortion is required Similarly in all cases there will be zero distortion required to fit the data to a primitive triclinic cell since all three unit cell lengths and angles are unrestricted The goal here is to find the highest symmetry lattice which fits the data with minimal distortion In the example above there are two candidates aside from the trivial case of P1 These are primitive monoclinic with a 0 18 distortion index and 21 THE H KL MANUAL The first crystal The last group of crystal Autoindex When Autoindexing Fails Se aa Utter failure experimental conditions 28 primitive orthorhombic with a 0 30 distortion index Since 0 30 is in this case an acceptable amount of distortion especially given that the unit cell parameters were refined from a single frame a reasonable choice for the lattice would be primitive orthorhombic Note that this choice of lattice must be confirmed since it may well be a monoclinic lattice that has a beta angle which is very close to 90 degrees To do this process the data with Denzo and then postrefine the unit cell parameters and scale the intensities in Scalep
148. en most of the peak search result may be explained with a smaller real space unit cell To prevent autoindexing from finding such a smaller cell one should enhance the fraction of odd reflections used in autoindexing by changing the peak search criteria or by using only high resolution reflections in autoindexing If one still cannot index odd reflections then one should consider ignoring them all together In such a case one can solve the structure in a smaller unit cell and the resulting structural error will not be very significant if odd reflections are much weaker than even ones means that a lattice other than P1 was predicted but it had a high distortion index say 1 1 5 This may not be a failure but simply an indication that further refinement of parameters is needed because the initial detector parameters were very far off Unfortunately misindexing may give similar results A quick check of the superposition of the preds on the reflections should tell you if the refinement is likely to succeed Note that refinement may converge to low y values even with misindexed crystals but eventually the problem will surface and you will have to index correctly agree perfectly with the peaks and no detector or diffraction artifacts are visible but data scale poorly and produce many outliers This problem may be a simple mistake in data processing like using the wrong file format or a non uniform exposure during crystal oscillation The n
149. ent stuff but it still just sits there Try Crtl D UNIX or Crtl Z VMS to terminate input into the program I tried to run Scalepack I get an error message Exec format error Wrong Architecture Any ideas You try to run program compiled for different operating system or OS version Investigate I got back from SSRL collecting data on the MAR I want to process data with Denzo as well on our Indigo and I was wondering if I need to swap bytes for INST INST and Denzo programs recognize swap bytes automatically How do I get the radial and angular offsets for SSRL scanner Do you have those numbers Sorry we do not know Ask SSRL staff xxx TERMINATING usr perutz1 wladek DENZO DENZO Received signal 10 SIGBUS Bus error An error from old SUN version Problem specific to SUN If you get such an error contact authors of the program I have an older version of Scalepack which I assume is fine to use Or is it It should work but it is not as good as the new one Can you produce a version of Scalepack that will take up to 1 000 000 reflections Yes now it is a standard version Should we have Y SCALE 1 0 for this MAR plate Yes however it is a default value We are looking at some Daresbury MAR data The rotation axis is horizontal and I thought that we always processed this with FILM rotation 90 0 or 90 0 but now we are all using 0 0 Is this because the convention has changed I am worried that the polarization correc
150. ent that the x ray beam is vertical the fact remains that so far all macromolecular crystallography is done with horizontal x ray beams on the surface of good old planet Earth and thus it is a viable option While Denzo does not use this convention XdisplayF does Hey life is messy The spindle beam system is the Denzo crystal and cassette orientation convention In this system the z axis is again parallel to the beam the x axis is parallel to the spindle axis and the y axis also termed the vertical axis FOr Want Of a better word is perpendicular to the spindle and beam axes In terms of crystal orientation rotations E otz would again denote rotations of the crystal around the beam axis E rotx would denote rotations of the crystal about the spindle axis and E roty would denote rotations about the axis perpendicular to the beam and the spindle REFINEMENT The advantage of this system is that it is very intuitive to the crystallographer In addition with Eulerian y values of 0 90 180 or 270 degrees the spindle and vertical axes are parallel to the x or y axes of the detector The main disadvantage of this system is that it is dependent on knowing the geometry of the camera While the crystal and cassette orientations follow the spindle beam convention for Denzo the beam box box printout spot margin film width and length in the Denzo log file follow the data convention Determining the values of FILM ROTATI
151. er and combination The initial indexing procedure is automatic and uses only a single still or oscillation image Partially recorded reflections are used in the positional refinement of the crystal orientations for high accuracy and computational stability In addition Denzo when combined with the companion program Scalepack accurately tracks and sums up partially recorded reflections over an essentially unlimited number of frames This provides the user with the option of collecting thin oscillation frames or extracting usable data from crystals of high mosaicity To Denzo a data set involves two separate steps The first is to determine the precise crystal and detector orientation parameters for a single image in the collected series The intrinsic crystal parameters include lattice type and unit cell dimensions The crystal orientation parameters describe the orientation of the reciprocal lattice with respect to the spindle beam and vertical axes of the camera The detector and X ray parameters include the wavelength the crystal to detector distance the precise coordinates of the direct beam the detector missetting angles and the internal scanner alignment parameters such as the non orthogonality of the scanner head motions or the scale of the y pixels to the x pixels of the scanned image The second step in Denzo ing a data set is to process all of the images in the series using the parameters determined from indexing one
152. er to decrement The middle mouse button is analogous to one frame advance or decrease Has some help menus Not very helpful yet To close click Help _ again except that one click results in only a Buttons in the Zoom Window This displays a set of cross sections through the zoom window so that you can see the 2 D profiles of the pixels Re centers the zoom display on the selected point Works in either the zoom window or the full image window Displays the information about a particular point in the zoom window in the green box in the lower and upper left corners of the zoom window If the frame has been indexed already by Denzo then when you click on a spot the h k 1 indexes of that spot will be displayed Otherwise the display only shows the position of the spot and its intensity Increases the magnification of the zoom window When you get to really high magnification the intensity of each pixel is written in that pixel However this is not always accurate and will be discussed under the Green Box window section below Decreases the magnification of the zoom window This is a toggle through three different displays The first default display shows a circle centered on the predicted reflection This circle is always 6 pixels in diameter and has nothing to do with the size of the spot This is the display you see if you don t hit the Int box The first time you hit the _Int box you will see the integration box
153. erence better to search at medium to low resolution derivative is new crystal so need new ESTI MATED ERROR we need this here because new data expected fraction of outliers in the data This is critical because the presence of be indicative of a potential derivative this removes overloaded reflections from statistics Overloads may cause large differences but are not indicative of a derivative partials in derivative will be summed data format of potential derivative only reindex if native and derivative have different indices See scenario 5 3 frames of a potential Hg derivative e g 3 x 1 degree oscillations not interested in sca file only the log Eb The goal here is to scale a few frames of a potential derivative with a native data set Here you only do a single round of refinement not the iterative cycles done in Scenario 1 because you are not doing any cycles of rejection In this case three frames of the derivative say as much as 3 degrees will be compared with the native to a resolution of 3 5 A B This scenario is only valid for a small wedge because otherwise you will be looking at RinergeS which result from reflections combined with the derivative data set as well as with those of the native and so the statistics will be less indicative We are only interested in the last page of the Scalepack log file where we will examine the y values and Rmerge Here are some general rules for assessing your
154. es Aee20 deo GATO 49520 34562 L0 Z42 Gela OaE Pegg DeF os22 00 ALe eat soL ela ULES Mero T62 4095 6 A330 BYA Se 8067 Ow hse Deeg L62 Las 3472590 426321 Abo 2413 Oebss O20 LeS tet 8163 7 S1420 425627 Lale Grrl 02190 LAS adie O AILT e000 ADe TeSZ22 O26 ZON Teo hei 2624s T 2865 ABe 22 845 Osos G22 Deseo ea OSO Gad B82 TSG oy DOG Oe S63 All reflections LISST Toe GA6 3 Oe 40S 0 096 OLL But there are no reflections with I o I less than 5 Shell I Sigma in resolution shells Lower Upper No of reflections with I Sigma less than limit limit O 1 2 3 5 10 20 gt 20 total hs re 0 0 aT O O O 0 So 243 L372 4655 6027 Zt 2a20 O O 0 O 9 466 2152 3283 5435 220 192 0 O 0 O 36 804 2870 1756 4626 Pro baro 0 O 0 O 93 1201 2770 494 3264 ae as em bere ot O 0 O 0 AAG POSS ei 80 1897 a AE 0 O 0 O 60 859 TOTS L 20G eS eee eS 0 O 0 O 118 481 aba 0 OLS 1 45 1 39 O 0 O 0 73 204 211 0 ZEL AIP 5S E S O O O 0 Gl L11 LIZ O KIZ Troos La29 0 0 O O 89 602 665 0 665 All hkl 0 0 O Oo PSO B24 L13363 10269 23632 DENZO output Q In the Scalepack output I get in the table for I o in resolution shells 0 reflections in the first five columns I o 0 5 i e I only get reflections with I o gt 5 I nowhere input a 5o cutoff and looking through the manual the only explanation I found was the keyword WEAK LEVEL which has a default value of 5 On the other hand as I understand it this only means that reflections lt Jo ar
155. es are centered on pixel boundaries circles are centered on precise values the integration circle ellipse and background square are in white if one clicks the left I think mouse button a second time one gets red circles ellipses Are these red circle ellipses the integration box or the background box Both of them Example lt Bad Detector Pathology I didn t get satisfactory result from Scalepack I would check following a Beam line b Shutter c Detector d Crystal size radiation sensitivity mosaicity etc e Consistency of indexing if data from more than one crystals are used in some space groups f Crystal isomorphism It is now clear that we do have shutter problems We have only recently discovered that ours apparently either opens late or closes intermittently I recently scaled a data set that had good Ryerge values co high normalized c values in Scalepack Is there something wrong or should I ignore the c values I know that I can boost the o values in Scalepack with the ERROR SCALE FACTOR I was just wondering whether it could be due to any other parameter in Denzo For example incorrect background definition or box size or anything like that Our spots are very elongated but data is strong and values from Denzo appear to be good I know that other data sets give values in the range 1 2 but this one is in the range 3 4 Does it get harder when the SPOT elliptical is more asymmetric Reprocessing wit
156. f8 1 OUTPUT FILE name use single quotes for safety you must specify the output file name OUTPUT FILE people myname LYSOZ sca This command allows you to correct a mistaken polarization monochromator value entered when you ran Denzo It saves you from the chore of reprocessing all of your original images once you have learned of a mistake This is mainly for synchrotron users who collect very high resolution data format POLARIZATION Denzo value corrected value default the polarization is read off the header of the x file example POLARIZATION Denzo 0 0 corrected 0 9 POSTREFINE format default example 132 Number of cycles of postrefinement do be done Works in conjunction with the FIT commands described above If you set the number of postrefinement cycles to zero then postrefinement will be skipped even if you have FI T commands in your control file POSTREFINE integer postrefinement is not done unless you specify it POSTREFINE 10 PRINT modifiers format defaults PROFILE TEST THE SCALEPACK KEYWORD DESCRIPTIONS This flag tells the program to print the results of the specified calculation after every refinement cycle Useful in the case where one scales non isomorphous data and wants to make the outlier list short by using the statement PRINT total chi2 200 PRINT single chi2 100 total ichiZ cutoff for printing poor reflections Total y for all measurements of a reflection Default 5 Single ch
157. finement Statistics After each GO command Denzo not only updates the pred display it also prints out the numerical summary of the refinement cycle The first part lists the refined parameters their final values and how much they changed in the last refinement cycle The last part lists the number of spots used in the refinement the y values for the x and y positions of the preds and the predicted decrease in the y values from the previous refinement cycle If things refined well the number of spots used in the refinement should be a significant fraction of the total number of spots on the image This number reflects the number of preds that line up with observed spots The y values represent the average ratio squared of the error in the fitting divided by the expected error A good refinement will have x values near 1 0 In the early stages of refinement these values will be high but towards the end they should be below 2 or so The predicted decrease is a measure of the convergence of the refinement When the predicted decrease is near zero further GO commands will have no added effect In other words the current parameters have been refined as far as possible The magnitude of the x values is not a critical test for the success of Denzo since they 40 represent only the comparison of the spatial differences between the observed and predicted reflections to an error model that is typically quite strict in the pr
158. fo scala does york output file has this information see above Is it possible to suppress this function of putting information at the end of the x file No it is not possible Current rotaprepa can skip this information On integration the output often has many reflections with negative os what have I done wrong Nothing or almost nothing e g resolution limit too high for MAR scanner Could you tell me if the Denzo format has changed greatly Rotaprep use to convert Denzo images to mtz images but it no longer works I want to compare CCP4 processing to Scalepack processing Just using the Icf batch or york output file does not do the trick Use current version of rotaprep from ccp4 with york output file I am still unable to go to Scala since the ROT information does not carry through rotaprep So far though Scalepack does the same as rotavata and agrovata Not surprising Check with Eleanor Dodson or Phil Evans HKLQ amp A Does Scalepack use the top 6 line of a x Denzo output file when it reads them in Yes The last column in a x file is it used by Scalepack at all No It appears that the intensity has already had the Lorentz polarization factor applied to it Is this true Yes Why doesn t the center of the integration box fall on top of the center of the predictions vellow cyan circles I found them to be offset towards the upper and lower edges of the Zoom box but more or less centered in the middle Box
159. frame in the series The Denzo user has full control over which parameters are refined in every cycle The strategy in refinement therefore is to start at low resolution adjusting just a few parameters and then to extend the refinement to a higher resolution where more parameters can be fitted Once a data set is processed with Denzo the output is ready to be scaled using the companion program Scalepack Together these programs make up a useful package for the reduction and merging of X ray diffraction data DATA COLLECTION PROCESS Data Collection Process A recent review of basic problems of practical crystallography is given in Jan Drenth s Principles of X ray Protein Crystallo graphy Springer Verlag 1994 ISBN 0 387 94091 X Every element of the data collection process must function close to its optimum in order for one to solve a macromolecular structure The sheer amount of collected data makes computer programs an inevitable intermediary between the researcher and the experiment The HKL package provides several levels of insight into the data at each stage of the measurement and data analysis process E Scalepack which provides statistics for the full data set E Denzo which provides numerical analysis of one oscillation image and E XdisplayF which presents data visually up to the single pixel level Different problems manifest themselves most clearly at different levels of analysis The traditional method
160. h Denzo rarely improves y dramatically unless you correct a major mistake in the processing 151 The HKL Manual 152 I am trying to scale some data which processed very nicely from 30cm mar at 9 6 I am trying to see why it doesn t scale well If I change the value of ERROR SCALE FACTOR should I try to get the final e column at the bottom of the output of Scalepack to be 1 or should they be equal to the value of the ERROR SCALE FACTOR I give At the moment it looks like Shell Lower Upper Average Average Norm Linear Square limit Angstrom I error Stats Chari R laO R Lac LOO 2 00 4 at ZUG Laa LA SSS Ae S56 00o OAS W634 3342 LAUZ SSS J839 1030 26 34080 UT Ue 127 ZsAZ 2099 59864 0 4943 1 SO 3560 IrIS 0092 Os 74 Den Ot AS AAA ee BALL 2690 9 4 465 O LOL Os 60 Ait Big Oe Zi OAO AGa VAE PA E Se O01 Os lS E Aigo or NOU L Stes Aa PASSA yo eer Odio Ore BAO el eo 2240 LSLS G 221404 ZUA Aah 2a So Oxon Oro Dewees ADAS ALTS LIG ost Zea Oss 0 276 Zea 2 S07 TIGA 2OC4 2 Ze Odie ano Be 0S SA 2207 SOD COUS ak gala AOD Da Oe AAO A All reflections 923888 4959 6 AOA To Sion Oso 2 QTA with ERROR SCALE FACTOR 3 0 and ESTIMATED ERROR 0 06 Things are pretty bad I d try with Scalepack with e g ESTIMATED ERROR 0 10 usually I use 0 06 and REJECTION PROBABILITY 0 01 A value for ERROR SCALE FACTOR Usually about 1 5 seems quite good but I guess in this case I should experiment a bit but if I need to have ERROR
161. i2 cutoff for printing poor reflections y for any single measurement of a reflection Default 3 5 correlation prints correlation matrix Default no correlation no correlation solution prints new scale and B factors Default no solution no solution shifts no shifts prints changes to scale and B factors PRINT modifier PRINT total chi2 PRINT single chi2 PRINT no correlation PRINT no solution PRINT no shifts New scale and B factors will always be printed if the number of fitted parameters exceeds 300 PROFILE TEST is a flag that can be turned on and off as desired This is the default Bad profiles can be used to reject a DCREDUCE summed Akl Valid for ARCHIVE file format only Opposite of NO PROFILE TEST PROFILES FITTED format default This flag tells Scalepack how the profiles were treated by the indexing program e g Denzo The two choices for profiles are fitted and summed Valid for denzo_ip Denzo_york1 and xengen urf formats Most of the time people fit profiles so this is the default and need not be specified in the command file PROFILES FITTED this is the default 133 The HKL Manual PROFILES SUMMED Opposition of PROFILES FITTED This is not the default and must be specified if true RECSQ Metric tensor description of the reciprocal space unit cell Why would you ever use this when you can specify the real space constants format REGS O valuel value2 value6 default same as U
162. idea to start by first refining those values you are least sure of and which are most critical These are usually the crystal rotations unit cell parameters and the beam positions In addition the resolution limits must be set and generally you start refining at a medium resolution and progress to a higher resolution in one or two steps So the first round of refinement would start with two commands RESOLUTION LIMITS 100 2 8 FIT crystal rotx roty rotz cell x beam y beam GO GO GO This tells the program to refine the crystal rotation angles the unit cell lengths and angles and the direct beam position for three cycles 3 GO statements using reflections between 100 and 2 8 A and applying the results of the first cycle to the second and so on In the ImageWindow you will see the preds move after each cycle to their new positions If things look like they are getting closer you can do more cycles simply by typing GO more times Once you specify the FIT operation as in FIT crystal rotx roty rotz these parameters will be refined every time you type G0 no matter what you have specified on the same line as GO For example let s say that after the 3 cycles of CRYSTAL rotx roty rotz refinement you decide to refine the X BEAM and Y BEAM positions using the command FIT x beam y beam GO GO G0 Now not only will the beam position be refined but simultaneously so will the CRYSTAL rotx roty rotz rotations Once FIT is on for a parameter it is
163. ignals the end of the Denzo job Placed at the end of the command file or used to exit from Denzo during manual fitting and indexing END OF PACK REDUNDANT ERROR 54 Signals the SECTOR argument to be incremented by 1 The CALCULATE GO statement does now the previous work of END OF PACK The keyword ERROR and its modifiers defines the model for the error in the measurement of the spot positions The default values should be adequate in most cases and you should not worry too much about them The error model does however directly affect the a values printed out at the end of each cycle of refinement Since this is usually the only check you can reasonably make as to the success of your data reduction in Denzo it is wise to know modifiers format defaults example ALPHABETICAL KEYWORD DESCRIPTIONS something about how these are calculated The names of the modifiers contributing to the error definition should not necessarily be taken literally since they are short hand notation for complicated effects on the accuracy of the measurement The model has been developed and black boxed i e given the Z O Good Housekeeping seal of approval to give a realistic estimate of the weight each spot should contribute to the refinement of the crystal and camera coordinates This model gives y values close to one for good non mosaic crystals For mosaic crystals either error positional needs to be increased or higher values than 1 of y7 shou
164. in 1 degree e film rotation exactly depends on detector and y value of spindle e crystal unit cell lengths within a couple of Angstroms cell angles within a few degrees e space group lattice type e oscillation range e effective mosaic spread of the reflections If in doubt enter in a higher number A good choice is 0 1 to 0 3 degrees for capillary mounts and 0 5 to 1 0 degrees for frozen crystals e Furthermore I will assume that the non 20 dependent cassette rotation parameters are near zero All of the non trivial parameters listed above can be refined if they are close enough and the crystal orientation is figured out to a good first approximation You will know you have successfully indexed your crystal when the predicted reflections preds line up on the observed diffraction spots It is indeed possible to manually index crystals whose lattice type and unit cell parameters are not known However this significantly increases the number of variables you will have to worry about The indexing proceeds in two steps The first is to determine the spindle and vertical axes Then the crystal rotation angles are determined Determining the spindle axis and vertical axis 78 The SPINDLE AXIS is the unit cell vector which is parallel to the axis of rotation of the camera spindle It is given in Akl notation so that non major axes can also be specified For example the axis 0 O 1 means that the c axis is parallel to the s
165. in Denzo with the opposite sign to the R axis definition for 20 DISTANCE is defined in Denzo as the distance between crystal and a point there beam intersects detector it is R axis distance cos 20 X BEAM and Y BEAM are the actual position of the beam on the detector during exposure at the current 20 not at 20 0 X BEAM and Y BEAM change with 20 mostly Y BEAM If the detector is aligned perfectly X BEAM does not change The Y BEAM in extreme cases may be outside the detector The correct value may be negative or more than 200 in such case The definition in Denzo is that the cassette detector center of rotation is the point where the beam intersects detector surface R axis 20 rotation definition makes the crystal at the center of detector rotations All options in Denzo including autoindexing work with any detector orientation Only the resolution display in XdisplayF is incorrect for detector with 20 non zero Absorption Correction Factors Q Do you have any information about the use of the air absorption card in Denzo The manual says that reflections are attenuated by e 2 78 but it says nothing about the angular dependence or the wavelength dependence both critical factors in determining the true effect of air absorption on reflection intensity The air absorbs equally in all directions The program calculates air path and corrects for the absorption in the air It is a very small correction C I think I ve answered my ow
166. in Denzo will then be much more precise Otherwise the gain from reprocessing 1s very small What does START REFINEMENT do Start a loop that finishes at CALCULATE G0 Loop range is defined by SECTOR 1 to 50 This will execute the loop 50 times 157 The HKL Manual C Zero o often comes from too small or absent PROFILE FITTING RADIUS Increase the value and judge it on the display click top left button For small molecule crystal spot density is smaller and larger PROFILE FITTING RADIUS may be optimal Rhombohedral Q R32 causes problems You can reindex to hexagonal setting in Scalepack See Scenario 5 of the Scalepack manual You can use both primitive and hexagonal settings in Scalepack you can also reindex primitive setting from Denzo to hexagonal in Scalepack C I have now processed my data as C2 monoclinic and the processing looks good but after discussing it I realized that C monoclinic lattice is a subset of rhombohedral lattice Twinned crystals The problem is that the data is from a twinned crystal Both orientation angles have been figured out manually for two lattices Due to the overlapping most of frames have relative high average value up to 6 0 sometimes The data set reduced turn out to be good enough to yield right molecular replacement answer without any further modification of the data reduced by Scalepack But for high resolution structure the data is bad as regards to R value 28 after simu
167. inal index and film number in free format This information is most easily edited from the printed output of a previous run can contain the whole line One can read it from a reject file with command reject format RE ECT HKL default this is not the default 134 THE SCALEPACK KEYWORD DESCRIPTIONS REJECT OUTLIERS Automatic rejection of outliers Not yet reliably implemented This is supposed to replace multiple rounds of running Scalepack The idea is that when this keyword is set Scalepack runs once makes a reject dat file then runs again reads the reject dat file and scales the data based on the reduced set of observations REJECTION PROBABILITY Applies Bayesian rejection of outliers Rejected observations are written to the reject dat file see WRITE REJECTION FILE The whole Akl all original hk s with the same unique Ak with at least one observation with a probability of being an outlier greater than or equal to the value specified after the WRITE REJECTION FILE keyword default 0 9 are written to the log file This is an estimation on your part of how frequently you expect any observation to be an outlier In principle the rejection probability should be about equal to the number of outliers divided by the number of observations A good value to use is 1 10 000 i e 0 0001 for normal good data If you have a non random signal in your background e g satellite crystals malfunctioning detector ice rings then you wil
168. indrical Weissenberg MAC Science DIP 320 Weissenberg scanner MAC Science DIP 320 Weissenberg scanner 16bit MAC Science DIP 3200 Weissenberg scanner MAC Science DIP 3200 2 Weissenberg scanner long MAC Science DIP 420 off line scanner INST_DIP2000 INST_DIP3000 INST_DIP320 MAC Science DIP scanner for neutrons 0 2mm pixel MAC Science DIP scanner for neutrons 0 4mm pixel ADSC Quantum4 CCD detector raw data ADSC Quantum4 CCD detector raw data dezingering on ccd brandeis 2k ccd unsupported b4 ccd bl19 ccd esrf fit2d transformed ccd brandeis 2k ccd unsupported b4 ccd bl19 ccd esrf fit2d transformed ccd esrf fit2d transformed wrong ccd esrf fit2d transformed wrong distance ccd esrf 1024 ccd esrf 1242 ccd chess 55mm ccd chess 80mm ccd chess 80mm binned dip 100 dip 2020 dip 2020b dip 2030 dip 2030b dip 2040b dip 3000 dip 320 dip 320b dip 3200 dip 3200 dip 420 dip nf dip nc ccd adsc quantum4 ccd adsc quantum4 dz distance ccd esrf 1024 ccd esrf 1242 ccd chess 55mm ccd chess 80mm ccd chess 80mm binned dip 100 dip 2020 dip 2020b dip 2030 dip 2030b dip 2040b dip 3000 dip 320 dip 320b dip 3200 dip 32001 dip 420 dip nf dip nc ccd adsc quantum4 ccd adsc quantum4 dz 83 THE HKL MANUAL Detector Description Modifier XdisplayF Ver 1 5 and lower Modifier XdisplayF ADSC Quantum4 CCD detector raw data binned ADSC Quantum4 CCD
169. ines that program authors may not be aware of I told you that we have a SGI Power Series with VGX Graphics This will handle your program correct Any special hints No hints You will not take advantage of VGX graphics over standard entry level graphics What have you done to Scalepack such that it now exceeds any page file quota which I have and which runs the older version quite happily Increased allocation of memory for storage of reflections Also are you still supporting your IPVIEW Denzo package for SGI machines We have noticed some binary executable incompatibilities between our platform and older ones which are meant to be compatible SGI creates hopeless number of almost compatible releases of operating system SGI are generally supported however older versions 4 0 5 or so may produce problems I m not sure what the byte swap of the image is Denzo and Scalepack are now educated enough about little endian and big endian computers so both programs can cope with byte swapping problems automatically On UNIX OSF computers you may install all programs in usr local bin directory remember that only root privileges allow you to do that Check if in your path statement usr local bin is included Another strategy is to put all files in arbitrary directory for example usr users wladek HKL and add to your cshrc file a bunch of aliases like alias denzo usr users wladek HKL denzo alias ccd usr users wladek HKL INST _CCD
170. insulting Rather than taking offense please read these few sections in the playful spirit in which they were written and remember that the primary THE HKL MANUAL HKL user support audience for whom this manual was written and at whom the barbs are directed is the author himself Questions about the programs described here should be directed to Zbyszek Otwinowski or Wladek Minor at the addresses below You may also use hk1l hkl xray com or news bionet xtallography as a medium for HKL user s ideas exchange HKL mailing list Maj ordomo hk1 hk1l xray com Information and user support hk1l hkl xray com HKL WWW homepage http www hkl xray com Zbyszek Otwinowski Zbyszek mix swmed edu Wladek Minor Wladek iwonka med virginia edu How to reference Denzo and HKL History Acknowledgments If you want to reference the HKL package for publication use the following citation Z Otwinowski and W Minor Processing of X ray Diffraction Data Collected in Oscillation Mode Methods in Enzymology Volume 276 Macromolecular Crystallography part A p 307 326 1997 C W Carter Jr amp R M Sweet Eds Academic Press The authors encourage referring to the use of the HKL programs when describing structural work This is particularly significant when statistics from data reduction are being quoted for example resolution limit signal to noise ratio Or Rmerge The HKL package when properly used gives a realistic estimate of mea
171. ion can Because you can get different solution related by lattice symmetry you must only do autoindexing once per crystal I would like to remind you that in order to find indexing you need to collect one oscillation image i e without Weissenberg motion and only later move to Weissenberg mode On a second data set from another protein I found that XDISPLAYF and Denzo were unable to index from the first oscillation which was very close to a zone i e it could not get the third axis Depending on the way I picked peaks Denzo either failed entirely with zeros in the matrix or it found a cell axis of 1 25 A as compared to 164 A If I chose a frame midway through the data collection then Denzo indexed quite quickly The problem was really quite different The oscillation range was too large for the combination of resolution and unit cell Lunes were overlapping and autoindexing failed by the way this applies to all autoindexing algorithms When you rotated the crystal you were no longer collecting data down 164A cell and lune overlap was less severe The right solution is to collect small oscillation sectors down long unit cell axes If you already have collected large oscillation sectors down the long unit cell restricting autoindexing to using only low resolution data may help In this case you may try RESOLUTION LIMITS 40 6 Is it possible to create a file of spots for Denzo that comes from more than one image We have a rather weak diff
172. ion range Only reflections that are single partials are included in the analysis A reflection in a negative zones means that the center of the reflection has already passed through the Bragg condition The fourth column lists the average observed partiality of the reflections in each zone and the histogram is just a plot of the average observed partiality In all cases no matter what mosaicity you have input to Denzo the histogram should pass through 50 in the zero zone If the mosaicity you guessed input was too small then the histogram will not descend to zero at mos 2 If the mosaicity you guessed input was too large then you will see a histogram which looks like a step function falling to zero just past the zero zone In the example below the mosaicity was chosen correctly 2 x 0 06 0 12 or slightly overestimated which is the preferred side to err on but there is some diffuse scattering which accounts for the small tail on the histogram Note also that this is a particularly nice example because there are many contributing reflections to each zone If there are only a few reflections the histogram may be quite choppy If the histogram does not resemble any of the shapes mentioned here it may be indicative of motor spindle or shutter problems refl zone av part 10 0 10 20 30 40 50 60 70 80 90 100 110 169 0 060 a 0 054 0 927 Axkkkkkkkkkkkkkkkkkkkkk kkkkkk kkkkkkkkkk 179 0 054 0 048 1 019 kkkkkkkkkkkkkkkkkkkk
173. is better to err on the side of a slightly high mosaicity because this way you can be sure of catching all the spots This is because Scalepack can re label measured partial reflections as non diffracting but cannot do the opposite i e make up a reflection which Denzo did not find Postrefinement with Scalepack will calculate an accurate measure of mosaicity and if the value used in Denzo turns out to be low or way too high the frames can be reprocessed with Denzo using this corrected value for the mosaicity When Denzo actually fits a profile to each reflection and measures its intensity it does so only to the portion of the reflection that is within the borders of the predicted spot the pred If you choose a pred which is E too small intensities will be too low because part of the reflection was not measured E too big you run into the risk of a rejecting reflections due to overlaps overlaps are defined by whether or not the predicted spots guard or background regions are overlapped not whether the actual spots are b including portions of a neighboring reflection in the measured intensity and 35 THE H KL MANUAL Profile Fitting Radius 36 c including portions of the background in your intensity measurement Therefore it is wise to choose a pred size that matches those of your reflections as closely as possible Often the diffracted spots are slightly elliptical They also vary in shape with the
174. ision of the summed intensities If the predictions do not match the peak position exactly one has to enlarge the expected spot area in order to sum the intensity of the whole spot This enlargement of the predicted spot area increases the total background to be subtracted A larger background has a larger variance and this adds to the measurement s variance Auto centering of the spot area can compensate for such errors in the prediction but this works well only for strong spots It would seriously bias the calculated intensity if applied individually to every spot Some programs do auto centering by averaging the local deviations between the observed and the predicted positions While this is not done explicitly in Denzo the profile prediction algorithm used in the program has a similar effect A correction for the non linear response function of the detector to the photon flux is applied internally in Denzo so that it can read the original data without the need for any prior transformations with the exception of the data from spiral scanners Pixel values can represent two special cases no measurement or detector overload Overloaded pixels are assumed to be close to the center of gravity of the diffraction spots and as such they are used in determining the spot centroids Pixels that are either overloaded or have no measurement are ignored in calculating the spot intensity by the profile fitting but the existence of such pixels in the spot a
175. it cell specified by your command Denzo bug I am very dissatisfied with the inability of Denzo to correctly predict and integrate low resolution reflections If one has a 3 mosaic crystal and you have to give it a 1 5 degree mosaicity in order to get a mere prediction of a low resolution reflection this is ridiculous Whoa fella The problem is with the wrong value of X BEAM Y BEAM and resulting misindexing of the image If all reflections have wrong index at most random crystal orientations low resolution reflections will not match predicted pattern With right beam position mosaicity 0 3 did generate all low and high resolution reflections Program is OK Actually I did manage to index the 9 deg rotations Autoindexing of 9 deg rotations is tricky and often fails You were lucky Is the reason that you recommend using at least twice as many pixels for background as for spot definition simply statistical Yes Is it possible to create spots masks of differing ellipticity for the same image plate No When I redefine the box parameters the shape of the spot ellipse changes also Why If you change box parameters you may change parity even of odd number of pixels in the box Spot ellipse has the same parity as box so there has to be some adjustment to its shape Is the definition of the x y directions in the averaged profile boxes printed by Denzo the same as in INST_HAMBURGD I have the impression that the axes are interchanged
176. ize and overlap towards larger or smaller spots You can also edit results of the peak search procedure by using Pick Add and Pick Remove options For example you can INDEXING remove a Series of peaks by clicking first Pick Remove then clicking in either zoom or main window on peaks you want to remove Ice rings There is no perfect number of spots to pick for autoindexing because it is the quality of Rn the picked peaks not the quantity which is critical The peaks used in autoindexing should come mostly say gt 90 from a single lattice and should line up on real reflections in the image If most do not then there is something wrong and you may have to change some parameters as discussed below under When Autoindexing Fails The two most common problems which interfere with the peak search are satellite crystals and ice rings With the noisy image the peak search program may pick spots which are clearly not part of the lattice By clicking on Fewer Peaks the number of peaks selected can be reduced If your diffraction pattern has some strong peaks these will usually get picked before the noise Ice rings pose a different problem because the peak search program will pick tons of spots within the very intense ice ring However it is easy to overcome this by simply specifying resolution limits say 20 to 4 1 A in the autoindexing macro file auto dat which exclude the ice ring more on this later in the ce Rings sectio
177. k to include this extra frame It only affects a very small fraction of the reflections Opposite of DO NOT EXTEND PARTIALS EXTEND PARTI ALS this is the default This specifies the files read by Scalepack The input has two components The first is a number The second is a file name which usually contains wildcard characters that are incremented by the SECTOR command The number which follows FILE specifies the starting batch number A batch previously called a film can be as small as a single x file or the equivalent It can be a group of x files even an entire data set The most frequent content of a batch is a single x file This conversion of files into batches is particularly useful if you want to scale more than one data set together For example let s say you want to scale 10 oscillation frames numbered 1 through 10 from first data set with 37 oscillation frames numbered 1 through 37 from other one The FILE statement will take each of the individual x files and assign them a batch number Thus you would enter something like this FILE 1 setone x and FILE 101 settwo x see Scenario 5d Thus batch numbers 1 10 will correspond to files setone001 x setone002 x etc assuming you used the SECTOR 1 to 10 command above FI LE so that the wildcards would be substituted with numbers The batch numbers 101 137 will correspond to files settwo001 x settwo002 x etc EP the is replaced by the sector argument
178. kkkkkkkkkkkkkkkkkkkkk 155 0 048 0 042 0 929 xkkkkkkkkkkkkkkkkkkkkk kkkkkkkkkkkkkkkk 161 0 042 0 036 0 897 Axkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk 164 0 036 0 030 1 005 kxkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk 165 0 030 0 024 1 039 xAxkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk 176 0 024 0 018 1 040 kAxkkkkkkkkkkkkkkkkkkkkkkkkkk kkkkkkkkkkkkkkkk 175 0 018 0 012 0 898 kkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk 176 0 012 0 006 0 700 kAkkkkkkkkkkkkkk kkkkkk kk kkkkkkkk 160 0 006 0 000 0 607 Axkkkkkkkkkkkkk kk kkkkkkkkkk 195 0 000 0 006 0 409 RK RIK RK kk i kk 192 0 006 0 012 0 217 k RRR RH 188 0 012 0 018 0 142 163 0 018 0 024 0 091 188 0 024 0 030 0 072 162 0 030 0 036 0 055 193 0 036 0 042 0 056 186 0 042 0 048 0 042 167 0 048 0 054 0 041 181 0 054 0 060 0 041 4 The table below gives a summary of the types of reflections measured from the image The first column is the type of reflection The next three columns sort the reflections by strength above the weak level saturated overflows and below the weak level Note that weak reflections are NOT ignored or omitted they are just not used in the calculation of the average profile Overlap tabulates the number of reflections that are flagged as overlapping depending on whether you specified overlap spot guard or box background overflow is the number of reflections with unmeasurable ba
179. l crystal or cubic crystal exposed in the proper orientation It is mostly for synchrotron beamline staff not ordinary users 114 SCENARIOS Scenario 9 High resolution low resolution data sets from one crystal This scenario is useful if you have made multiple passes of data collection from the same crystal where one of the passes had short exposures to keep the low resolution reflections from saturating and the other pass was a longer exposure to get the high resolution data In the high resolution long exposure pass the low resolution reflections are mostly saturated Note the more complicated batch numbering scheme This is often typical of data collected at synchrotrons where there are gaps in the file numbering for various reasons This is particularly applicable to CCD detectors at CHESS scalehilo com NUMBER OF ZONES 10 ESTIMATED ERROR 0 04 0 04 0 04 0 04 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 5 REJECTION PROBABILITY 1 e 4 WRITE REJECTION FILE 0 5 SPACE GROUP P21 UNIT CELL 50 93 95 90 69 9 90 RESOLUTION 1 4 REFERENCE BATCH 1 reject POSTREFINE 20 FIT crystal a 0 to 1039 FIT crystal b 0 to 1039 FIT crystal c 0 to 1039 FIT crystal beta 0 to 1039 FIT batch rotz 0 to 1039 FIT batch roty 0 to 1039 FIT batch mosaicity 0 to 1039 ADD PARTIALS 0 to 180 184 to 216 Zi TO 230 LOOU To 1033 FORMAT denzo_ip GNORE OVERLOADS SECTOR O to 35 FILE O my_crystal_hi x SECTOR 39 to 18
180. l probably have to increase the rejection probability If you do not want to generate a reject list in the log file then omit the keyword A comparison of R4jinear and R squarea 1S Often helpful for deciding whether to increase the rejection probability See the discussion following Scenario I for more on this The rejection algorithm used in Scalepack is the most sensible and statistically sound outlier rejection algorithm Unlike some other programs the Scalepack outlier rejection is based on comparing differences to o So it will typically reject say 4 or 5 o outliers Some reflections with a large discrepancy from the average value of I may simply represent a lack of adequate statistics in measurement and not a mistake non random error in measurement If a reflection has a large discrepancy from the average and a large o its contribution to the average will be very small anyway Here s an example of how the outlier rejection algorithm works Se e 0 1 2 10 1 3 ee 4 Although you may think that 5 is the outlier in fact the consistent set of observations will be 1 2 3 5 and 4 will be the outlier Why is this A consistent average will be 10 0 or more exactly 10 000003 because observation 5 has a 1 1000000 lower statistical weight than the other observations due to its large o Observation 4 on the other hand which was measured more accurately will be 8 7 o deviations from the average of the remaining observatio
181. l values This is done in a procedure referred to as a global refinement or postrefinement The implementation of this method in Scalepack allows for separate refinement of the orientation of each image but with the same unit cell value for the whole data set In each batch of data a batch is typically one image different unit cell parameters may be poorly determined However in a typical data set there are enough orientations to determine precisely all unit cell lengths and angles Global refinement is also more precise than the processing of a single image in the determination of crystal mosaicity and the orientation of each image Scalepack allows you to analyze the quality of the diffraction data from your crystals The quality of the diffraction data is judged by many standards ultimately including the quality of the electron density map which results but a widely used objective standard is based on the statistics which result from scaling and merging the raw diffraction data This program calculates single isotropic scale and B factors for each of the films or frames of processed data that are input Scaled merged data are output A multiplicative correction factor ere 20 sin 2 A Ses scale is applied to I and o 1 Merging of redundant measurements which occurs throughout the refinement process is done as follows The multiplicative factor for each measurement is calculated from the scale and B of its film an
182. last refined position unless you re enter them manually and will not refine on subsequent GO commands unless you specify FIT again If the autoindexing succeeded subsequent refinement never should make things worse If it does it is best to make sure that the autoindexing really did work after all Go back and re read the section on failures of autoindexing and try the suggestions given there This is probably easier than trying to index your crystal manually As a general guide for automatic scanners the parameters are refined in this order CRYSTAL rotx roty rotz X BEAM Y BEAM UNIT CELL CROSSFIRE x y xy CASSETTE rotx roty RADI AL OFFSET and ANGULAR OFFSET for radial scanners Y SCALE and SKEW for orthogonal scanners DI STANCE Not all parameters are refined for every crystal because some are highly correlated In particular the crystal to film distance and the detector parameters Y SCALE and SKEW or RADIAL OFFSET are correlated with the unit cell dimensions A particularly dangerous combination to be aware of is the simultaneous refinement of DI STANCE CELL and RADI AL OFFSET These are so highly correlated that they can be refined only simultaneously for high quality high resolution data Generally the greater the curvature of the Ewald sphere and the higher the symmetry of the crystal lattice the better conditioned the problem That is correlations are weaker and parameters are more independent Some simple rules for when
183. lated annealing So I do want to get a better data set If one could modify the source code a little bit it will do the work The idea is 1 Do a prediction of spots from one lattice 2 Pull out a box around each predicted spot set the background to 0 or averaged background value of the frame 3 Do everything else the same as original Denzo do for the reflections from the other lattice I tried to do it myself But it is really difficult for me to decode the Denzo Would you like to tell me if that is possible for you to do it for me or is there any help that I can get from you The algorithm needs to be somewhat more complicated One has to include the possibilities of overlap between lattices Maybe in the future Q All I need is a way of being able to process data from a twinned crystal and I d be so happy Put detector as far back as possible to separate the two lattices Process one at the time Hope for best Do not expect outstanding results 158 INDEX e 9 semicolon Denzo keyword 71 input Denzo keyword 59 input Scalepack keyword 129 Z 20 angle beam to detector axis 28 53 142 20 angle beam to detector axis 23 54 2 D profiles of the spots displaying 20 A A D test XdisplayF button 19 absorption 8 31 113 144 AIR ABSORPTION LENGTH Denzo keyword 51 correction factors 142 ADD PARTI ALS Scalepack keyword 121 ADD Scalepack keyword 121 advantages of using Denzo 6 AIR ABSORPTI
184. ld be expected positional density overload y x factor The minimum error in mm in the positional measurement It is usually a fraction of the pixel size typically one half or one quarter of a pixel Its squared value is added to the other components of the error estimate Enlarging this value will lower your y values but will not necessarily improve your data if you have to increase this value much for a well tested instrument like MacScience or R axis check the instrument alignment Models the error measurement of the optical density of the film The error equals the square root of the corrected reading times the density value This is used only to estimate the error in spot positions and in the definition of strong reflections the error density is affected by the voltage on the photomultiplier The higher the voltage the higher the error density with a very steep dependence Refers to the increase in the modeled positional error due to the existence of pixels with intensity overflow For every overflowed pixel the modeled error squared is increased by the overload value squared This reflects the increase in the positional error of the spot due to its being close to the oscillation axis The error estimate is by 1 the ratio of y x factor to the absolute value of the tangent of the angle between the oscillation axis and the line connecting the spot with the beam spot This prevents the spot distortion which occurs
185. le Let s define the relative change in the square of the reflection width gt f w A w 5 REFINEMENT Diamond calculated that the fitted intensity will be wrong by a factor of 6 The averaging of profiles adds a value of r where r is the detector pixel size to the value of f Averaging will increase the profile fitted intensity of most reflections by a constant multiplicative factor which has little effect on crystallographic procedures The interpolation broadens the profile by a factor dependent upon the position of the predicted reflection relative to the pixel boundaries The interpolation will also increase f by a number between zero and r The interpolation method will increase the profile fitted intensities on average by the same factor but will also add random noise to the reduced data Other aspects of spot integration Ice Rings There are other often subtle ways in which errors in spot positions can lead to serious integration errors In many experiments the detector is placed as close as possible to the crystal while keeping the diffraction spots separated In such cases the reflections are barely separated Even small errors in spot prediction would make integration and background measurement areas of a reflection intrude upon the adjacent peaks and thus would lead to an inaccurate estimation of the peaks intensities Errors in the prediction of spot positions also affect the statistical error prec
186. les vary a lot across the film then you would choose a smaller radius If you have a small lattice and the spots are widely separated you might want to choose a larger radius E The calculation of the average profile is a time consuming task proportional to the number of format default example QUIT RADIAL OFFSET 66 spots in the profile fitting radius circle but too small a radius will not capture enough spots and lead to a noisy average profile Too large a radius will average out significant profile variations To see the PROFILE FITTING RADIUS in ImageWindow of XdisplayF click Prof Fit R A thin white circle corresponding to the profile fitting radius will appear and will move around with the cursor only those spots whose intensity is above the weak level will have predictions displayed in this mode and only these reflections are used in the calculation of the average spot profile to return to the normal mode click again A good rule of thumb is that the PROFILE FITTING RADIUS should enclose about 10 to 50 spots whose intensities are above the weak level Useful increments of the profile fitting radius are 2 5 of the detector size e g 5 mm for a 200 mm wide IP PROFILE FITTING RADIUS value_ mm 2040 PROFILE FITTING RADIUS 30 0 O The circles in the upper left quadrant are two 20 mm profile fitting radii centered on adjacent spots Each of the hundreds of spots on the film would have its own radius on a
187. lose to or somewhat larger say 50 than what you really suspect is the longest vector in the unit cell format LONGEST VECTOR value 7 A distance default LONGEST VECTOR 2 spot radius example LONGEST VECTOR 300 60 ALPHABETICAL KEYWORD DESCRIPTIONS MONOCHROMATOR Defines the polarization of the X ray beam Same as the keyword POLARIZATION This correction is not a big deal except for synchrotron radiation where it is a minor deal unless you are collecting high angle data The Denzo definition of polarization is as follows 1 1 where Z is the intensity of the electric field squared of the X ray beam along the spindle axis and J is the intensity perpendicular to the spindle axis Values range from 1 0 to 1 0 Generally speaking this is not a number you can measure very easily and is often calculated theoretically It is really the responsibility of beam line operators to know this value for their beam lines Note that the sign of the correction depends on whether the spindle is horizontal or vertical As a rule of thumb if the polarization correction is gt 0 then the intensity of the X ray scattering is stronger perpendicular to the axis of the spindle than it is to the axis parallel to the spindle You can check the background of your image to confirm this Typical values range from 0 11 for graphite to 0 9 for synchrotron beams the spindle axis is typically oriented at synchrotron
188. m is more complicated or you have data from a non automatic scanner you may have to read the rest of this manual Basically you will be using as many defaults as possible Hopefully someone has installed the programs on your computer and set up the proper aliases Step 1 Load your frames onto a local disk for fast access Put all your files in the same directory Make sure you have the four processing command files called site dat myexperiment dat auto dat and refine dat Site dat should have been set up for your X ray and detector system see Getting that First Orientation Indexing Step 2 Sit down at your fast computer workstation e g Alpha Indigo etc Open two windows I call them Image Window and Denzo Window and in each window set your directory to the directory where you are going to store the results from your processing Step 3 Edit the file myexperiment dat to update the experimental parameters distance oscillation range and start space group or lattice sector range mosaicity and file names If you don t know your space group or lattice enter P1 there Edit file refine dat to include the refinement resolution ranges Step 4 ImageWindow xdisp lt format gt xtal osc 1 see appendix Defector launches the display program Specifications DenzoWindow denzo launches Denzo get a blank line prompt Step 5 ImageWindow Click on 0 Peak Sear See tiny circles Denzo Window site dat myexperimen
189. metimes editing of weaker reflections and resolution cuts can make one crystal dominate the peak search list enough for the autoindexing method to succeed If crystals have a similar orientation sometimes using only very low resolution data can be the right method In case of twinned crystals autoindexing sometimes finds a superlattice that finds integer indexes for both crystals In such cases Denzo solves the problem of finding the best 3 d lattice that goes through all of the observed peaks Unfortunately for a twinned crystal this is a mathematically correct solution to the wrong problem Since there is no general algorithm to index a diffraction image from multiple crystals one has to rely on the human ability to sort out complex patterns in such cases The approximate orientation can be determined by an iterative trial and error process where the predicted pattern is being adjusted keeping the diffraction image constant A crystal orientation can be defined relative to any principal or higher order zone perpendicular to the X ray beam This flexibility helps the interactive indexing when only a higher order zone is visible in the diffraction pattern This is particularly useful in centered space groups where it is easier to orient a diagonal zone rather than a major one Manipulation of the predicted diffraction patterns also can be used to simulate diffraction experiments The simulation can help set a proper data collection strategy in order to av
190. mewhat off so the refinement does not kick in I have 60 frames but the data is weak and only to 4A There are some improvements to refinement algorithm that make it more stable however the problem is likely be somewhere else One likely explanation that would agree perfectly with your description is misindexing of lattice by one index The solution in such a case is to input correct or close to correct values of X BEAM and Y BEAM Is it the SECTOR n to m command that sets up the do loop for data integration Therefore if one puts this command at the beginning of the auto dat section one will get independent indexing of each image if it goes at the beginning of refine dat then one indexing is used and each image is simply made to fit as best it can Right No It does not matter where you put the SECTOR n to mcommand as long as it is before the first GO statement The do loop starts at START REFINEMENT and ends at CALCULATE GO When I ran a batch job to process 46 frames I noticed that the CRYSTAL rotx was not updated with the oscillation step during the time of each job For example it I start with CRYSTAL rotx of 32 167 after 5 frames are processed the value of the CRYSTAL rotx is about the same 31 469 However the OSCILLATION start and end are updated each time How can we understand this That you collected data starting each image at a different OSCILLATION start the way everybody collects his her data Last time you said that CR
191. minor_ mm value angle_ degrees default a ring just outside of the spot border examples BACKGROUND radius 0 45 5I THE HKL MANUAL BAD VALUE BOX E modifier formats default example CALCULATE CASSETTE modifier formats default example BACKGROUND elliptical 0 80 0 70 0 0 The value in the image file which does not represent a measurement Typically 0 When set to 40000 it disables 0 Used only on DIP series scanners to process small molecule data Defines the size of the box where both the spot and the background are measured given as two lengths in mm If the BOX size is larger than the average separation between spots 1 e the boxes overlap then the program will be slowed down by the spot overlap calculation The current limit on the box size is 90 pixels in each direction for a total of 8100 pixels The box size can alternatively be specified by the BOX where the size is given in pixels instead of mm print tells the program to print the spot definition at the beginning of every refinement cycle 1 e for every G0 statement BOX value1_ mm slow scan direction value2_ mm fast scan direction BOX print valuel value2 none BOX 2 4 2 4 This command tells the program that it is the last cycle when the spot intensities are calculated and written to the film output file It is followed by the command GO After the cycle both the RAW DATA FILE and the FILM OUTPUT FILE are closed Signals
192. n The picked peaks are written to a file called peaks file in whatever directory you happened to launch the display program This file is now read by Denzo to autoindex the image Running autoindexing Back to the DenzoWindow now you are ready to input the initial parameters and start the autoindexing For this purpose you will need the following files the oscillation image file the peaks file file and three macro files called site dat myexperiment dat and auto dat These three macros list the detector and crystal parameters and tell the program to begin the autoindexing Then in your DenzoWindow you would simply type site dat myexperiment dat and auto dat and the program begins Here s an example of the macro files Site dat FORMAT dip2000 detector type Very important WAVELENGTH 1 5418 FILM ROTATION 270 00 careful only for certain goniostats CASSETTE roty 0 465 rotx 0 102 probably determined by previous user ERROR density 0 60 error parameters change only if it has been determined accurately to be different from default BOX 2 4 2 4 SPOT ELLIPTICAL 0 6 0 7 spot shape descriptions etc BACKGROUND elliptical 0 7 0 8 OVERLAP spot flags overlapped spots X BEAM 99 4 Very important detector parameters Y BEAM 99 6 In this example and throughout this manual each major keyword is in BOLD CAPITALS modifiers and values are printed in lower case letters Comments are italicized and do not actually appear in the c
193. n CROSSFIRE x y xy FIT KEYWORD modifier all keyword parameters fixed i e type GO and nothing will be refined FIT CRYSTAL rotx roty rotz FIT X BEAM Y BEAM Y SCALE SKEW DISTANCE GO GO GO sT THE HKL MANUAL gt Do not run alternative cycles of fit and fix commands This is the old way and does not take advantage of the fact that the Denzo refinement is very stable and can refine many parameters simultaneously Simultaneous refinement is the best way to avoid getting into a false minima FORMAT A critical keyword The modifier following this keyword tells the program about a series of defaults to use for the detector parameters This keyword can thus substitute for many of the parameters called by for example the FI LM or MAGI NG PLATE keyword A detailed table of all the keywords specified by the format keyword can be found in the Defaults Values and Detector Specific Information appendixes modifiers see Detector Specific Information appendix format FORMAT modifier default this is a required input example FORMAT dip2000 GO This command tells the program to execute one iteration cycle of refinement After the cycle the fitted parameters will have new values You can specify more than one GO in one line This will tell Denzo to print results of the fit only for the last cycle from the series of GOs See also FIT FIX IBOX REDUNDANT BOX is exactly the same as the keyword BOX except that instead of giving the dimensions
194. n smaller values can be adjusted if you have an ice ring which interferes with the indexing you can set the resolution limits to say 100 4 1 A to eliminate these reflections a small satellite crystal which contributes some reflections at low resolution but not at high resolution you can set the resolution limits to say 8 2 A thereby excluding the strongest contributors from the contaminating lattice a cracked crystal where the two lattices are only 2 3 degrees apart you can often adjust the resolution limits as in the case of an ice ring so that you index at low resolution where the two lattices are approximately identical and then use manual refinement to zero in on both lattices The idea here is to index each lattice separately and to combine the data later in Scalepack a twinned crystal where the two lattices have a non random orientation relative to one another the program may fail spectacularly by finding a superlattice which indexes both lattices simultaneously but cannot be used to integrate the data because the superlattice does not usually fit both lattices exactly Then it s time to call an expert or grow new crystals If you have a number of spurious peaks in the image distributed randomly over the image you can adjust the value of the WEAK LEVEL which appears before the first GO statement in the auto dat macro This will affect the number of peaks accepted by the autoindexing routine By increasing the WEAK LEV
195. n be of several types Some options are printed by default at every cycle some are printed on the last cycle of every film of the refinement and some are never used unless specified See specifically the various print descriptions below chi square print the y statistics including the number of spots fitted the y of x and y a measure of how well the spots were fitted based on what sort of error you expected see also ERROR and the predicted decrease in the ys after the current cycle of refinement For those of you who just want numbers y7 s should be close to 1 0 between 0 5 and 2 0 and the predicted decrease should be near zero showing that the refinement has converged PLETE Print to the log file the observed and fitted profiles of those spots having the specified Akl value No density readings in the background will be printed By default no fitted spots printed hkl Print to the log file the density readings and other statistics of those spots having the specified Akl value By default no Akls printed memory Print statistics of the usage of variables profiles The profiles modifier tells the program to print the average spot profile in a specified number of regions to the log file It is best to refrain from printing profiles until the very last cycle of refinement or else your log file will become truly gargantuan You can also specify PRINT no profiles at the beginning of your refinement just to make sure it is off u
196. n question about AIR ABSORPTION LENGTH appropriate for 0 9 A synchrotron radiation It works out to 4500 mm using appendix 3 in Helliwell s book and the CRC Handbook of Chemistry amp Physics Cc With regard to our discussion of air absorption I looked up what is available in the International Tables and derived the equation log mu 3 014 log lambda 0 498 from some of the tabulated values This gives the following values for input to Denzo lambda air absorption length 590 mm 854 mm Denzo default is 860 928 mm 3145 mm Auto editin Q The program seems to reject most of the bad reflections that have interference from ice But still I see big peaks in the R factor distribution printed in Scalepack where I see the ice rings to be What s the best way to take care of ice rings It is done automatically however one can adjust REJ ECT fraction parameter to improve the accuracy of the rejections Default value is 0 75 increasing it to 0 8 0 85 may help Look at the display to see how many reflections turn red Q We got really bad icing on a couple of frames How do we leave them out You can omit whole frames from scaling or use RESOLUTION LIMITS to avoid ice In the extreme case you may reprocess the same image twice with resolution ranges below and above the ice rings 142 Autoindexing Q HKLQ amp A How does the definition of BOX and SPOT size in Denzo affect autoindexing The size of the integration spot aff
197. ne dat xbeam ybeam or manually format distance cassette auto dat verify enough preds hklpredictions to match spots overlaps amp others adjust mosaicity longest vector resolution limits weak level box and spot error density refine dat m edit peaks 1 Display the image and start Denzo adjustspot size checkprofile fit screen output or log file 2 Autoindex to deduce the orientation of the crystal and refine the parameters for the image you have displayed 3 The entire series of frames is processed either in batch or interactively 4 The log file of the batch job is then examined to see the results of the processing or the x files are displayed over the diffraction image or the data are processed in Scalepack What follows is a description of each of these operations 16 A DENZO SESSION Displaying the Image Using XdisplayF In the process which leads to A k l F and o the final statistics are affected by many factors The program outputs a substantial amount of information to help identify the potentially sub optimal steps in the data collection and reduction Interactions between factors like detector geometry crystal symmetry radiation damage detector failures non ideal crystals etc are far too complex to be formalized in a method that could be implemented strictly in a computer algorithm XdisplayF allows the crystallographer to visualize the data in th
198. ng direction but rather how the detector writes the data stream into the data file which thus makes the computer record of the exposed image Closer inspection of Figure 5 above should convince you that there are two possible hands to 47 THE AKL MANUAL Spiral scanners the conventions These are grouped as 1 4 5 and 8 and 2 3 6 and 7 To use an X ray film analogy this is equivalent to looking at the film either from the back side 1 e towards the X ray source or from the crystal side Denzo handles this aberration by changing the sign of the Y SCALE keyword from positive to negative If this is not done then the assignment of the Friedel mates Bijvoet differences will be inverted and thus the hand of the structure will be similarly inverted Obviously spiral scanners like the Mar scanner or some of the Mac Science scanners do not scan the IP in an orthogonal pattern However the controllers for these detectors convert the spiral data into Cartesian coordinates by using an interpolation algorithm The conventions listed in the table above for these detectors are based on how the data is written in the file Camera Conventions Gravity beam system Spindle beam system 48 It is inconvenient and non intuitive to work with images in the data conventions because we are used to thinking about diffraction geometry in terms of the x ray beam the spindle axis and the laboratory frame of reference horizontal vertical
199. ng done we aren t interested in sca only stats in log OUTPUT FILE NL junk sca we will be reading in a sca file from step 1 above FORMAT Scalepack FILE 1 mydatano sca Step 2 reads in the sca file which contains both I and I Obviously for centric reflections there is no I so the merging statistics will only reflect the non centric reflections You can tell what percentage of your data is being used to calculate the merging statistics by examining the redundancy table near the end of the log file Under the column of redundancy gt 2 you will find out what percentage of the data is being compared Since you only have I and I you will never have a redundancy of more than 2 The presence of an anomalous signal is detected by examining the y values in the log file from Step 2 Assuming that the errors in the Step 1 cycles were reasonable a y 1 in the log file of Step 2 means that there is no useful anomalous signal in your data On the other hand if y gt 1 in Step 2 this may indicate the presence of an anomalous signal Of course it may also indicate absorption problems detector problems etc But at the very least you can examine the x as a function of resolution and decide where to cut off your resolution for the purposes of calculating an anomalous difference Patterson use the shells where y gt 2 This whole analysis assumes that the error model is reasonable and gives you a y close to 1 when
200. nother way to picture these rotations is as the rotations of the arcs on a goniometer head mounted on the spindle of a camera CRYSTAL roty CRYSTAL rotx CRYSTAL rotz CRYSTAL roty CRYSTAL rotx S CRESTAL Ol Figure 3 Another way of thinking about Denzo rotation conventions arcs on a goniometer Tip to interpret this diagram for non horizontal spindles simply align the axis of your spindle with that of the spindle in the diagram All the rotations line up then as well It is easiest to change the crystal rotations one at a time in the following order CRYSTAL rotx spindle CRYSTAL roty and then CRYSTAL rotz To enter a new value type it in the command window as CRYSTAL rotx 22 0 G0 You can change more than one value at a time for example CRYSTAL rotx 20 rotz 5 0 roty 5 5 GO In the manual part of the indexing it is best to change the rotations in increments of no less than 0 5 degrees since finer adjustments can be done automatically by the program Denzo can recapitulate the current parameters with the command LI ST The best advice is to think clearly keep track of your movements and their results and be patient If you do manage to get a reasonable agreement with the diffraction pattern then proceed with refinement 80 ABOUT THE SCALEPACK MANUAL Film Cassette and Off line Image Plate Users The success of indexing with Denzo or any program for that matter depends critically on knowing or finding the
201. ns the biggest problem lies in keeping the detector flat during exposure and subsequent scanning In the manual systems it is much harder to model the possible departures from ideal flat or cylindrical geometry and Denzo makes only limited attempts to correct such distortions Non ideal film or IP geometry is one of the main factors behind the variable quality of data collected with the manual systems The main properties of the X ray beam that need to be checked are stability focus angular spread and wavelength in the case of MAD experiments Large fluctuations in beam intensity produce a variable background intensity and variable scale factors during scaling The quality of the beam focus is immediately visible in the spot profile of low resolution reflections The angular spread of the beam contributes to the reflection width and it may introduce overlaps between reflections for crystals with very long unit cells The beam properties except stability are best analyzed by the inspection of images The beam parameters are less significant for crystals with large mosaicity THE HKL MANUAL Summary 10 Macromolecular crystallography is an iterative process Rarely do the first crystals provide all the necessary data to solve the biological problem being studied Each step benefits from the experience learned in previous steps To monitor progress the HKL package provides two tools a Statistics both weighted yo and unweighted
202. ns because the expected error of the difference 4 lt 1 2 3 5 gt is 0 115 and 8 7 11 0 10 000003 0 115 So the observation with the largest deviation from the average is not necessarily a statistical outlier 135 The HKL Manual Rinerge after outlier rejection will be very bad 25 even if the average is measured extremely well with a sigma of 0 057 This has to do with R merge being an unweighted statistic and I o being a weighted statistic If some other program rejected observation 5 for no good reason and observation 4 for good reason R merge Would be 0 067 Practice shows that most programs would reject observation 5 because many users want a good R merge And sometimes these program would not even flag the true outlier observation 4 format REJECTION PROBABILITY value default 0 0001 example REJECTION PROBABILITY 0 0001 RESOLUTION Minimum d spacing for this run Default is the maximum resolution found in the input data One can supply two numbers in any order and they will be minimum and maximum d spacings format RESOLUTION value value default highest resolution detected in x file examples RESOLUTION 10 2 2 RESOLUTION 2 2 RESOLUTION STEP Changes the number of reflections per shell for the purposes of printing out statistics Formally it represents the exponent of the zone volume calculation Normally this is 3 because the volume of a sphere goes as radius cubed so all the statistics shell
203. ntil you want to turn it on shitts PRINT shifts tells the program to print the value the change from the previous value and the estimated error of all the keyworded parameters fitted on a current cycle of refinement PRINT no shifts turns this off solver This little used command prints out correlation matrices calculated in the refinement Statistics Print the hkl statistics including the number of whole and partially recorded reflections matched with preds Zones format PRINT PRINT PRINT PRINT PRINT ALPHABETICAL KEYWORD DESCRIPTIONS Print out a list of equivalent zones given as alternative values for CRYSTAL rota roty and rota One advantage to invoking this option is to use rot x roty and rotz values that are more convenient to remember or type For example instead of using crystal rot values of say 213 6 137 4 13 27 you could use the equivalent values of say I m just making these up 18 2 15 4 63 9 If you wanted to orient your crystal for a particular data collection scheme these alternative zones could make life easier PRINT zones lists all of these equivalent values chi square no chi square fitted integer integer integer hkl integer integer integer memory no memory profiles integer integer the first integer specifies the number of equal sized horizontal regions to divide the frame into the second specifies the number of equal sized vertical regions The total number of regions is the produ
204. o 2 1A I think though because it doesn t scale I will not use the data and instead collect it all again It seems like a waste but I think it is safer to start off with good data with good scaling etc before making maps and drawing conclusions from their interpretation Do you think I should give up with the data No other choice It seems you got screwed up HKLQ amp A Ignore Q I would like to be able to blank out the region of the imaging plate blocked by the beam stop I have tried IGNORE circle centre x centre _y radius and that works to get no predictions in a circular area Is there an equivalent ignore rectangle command Use GNORE circle 96 98 8 ignore circle of radius 8mm centered at 96 98 IGNORE quadrilateral 199 4 61 2 108 8 98 4 sR OSS ace A 199 4 74 8 These are the four corners of quadrilateral to be ignored Note that order of the 4 corners matter they have to be input in counterclockwise order C We ve discovered the utility of the IGNORE commands in Denzo to really define the beam stop shadow The data is weak enough that reflections behind the beam stop don t always get rejected in Scalepack The ignore circle command works fine Very handy Q The IGNORE bar also works but gives either a vertical bar R axis or something that looks like a very thin line at an angle The thin line is only detectable because the pattern of overlapped reflections is altered Can this bar be made horizontal Do not use GNOR
205. of statistics The difference between these numbers tells you what contribution statistical error makes to the total error o If the difference is small then reasonable changes in the values following the ESTI MATED ERROR keyword will not help your Xx much This is because the estimated errors represent your guess knowledge of the contribution of systematic error to the total error and a small difference indicates that estimated error is not contributing much If the difference between total and statistical error is significant and the X are far from 1 then consider adjusting the ESTI MATED ERROR values in the affected resolution shells 95 The HKL Manual 96 2 Examine the difference between the linear and square Rmerges For this comparison to be valid you cannot have strong overloads in your data so the command GNORE OVERLOADS must be given for this analysis If R is larger than R then the distribution of the deviations I I is skewed This implies that you have too many outliers in your data so you should reject more reflections The way to increase rejections is to choose a higher value of REJ ECTI ON PROBABILITY or write rejections to the reject file which have a lower probability of being outliers Useful increments are factors of 10 in REJECTION PROBABILITY and replacing default value of WRITE REJECTION FILE 0 9 with 0 5 If after one round this does not bring R close to R e g R is more than 50 greater than R
206. of the box in mm the dimensions of the box are given in pixels The in IBOX stands for integer The maximum dimensions of the box are 90 pixels in each direction format B0X integer integer_ pixels default none example IBOX 25 25 IGNORE Tells the program to ignore a specific region of the X ray image For example this is useful for excluding the shadow due to the beam stop Don t try to use this for masking out ice rings That is accomplished using the REJECT fraction keyword It is useful to think of GNORE as describing a masked out area of the film The mask is provided by a series of geometrical descriptions followed by their coordinates For example the beam stop shadow could be masked out by a combination of a quadrilateral and a circle i e a lollipop shape modifiers Quadrilateral A quadrilateral as defined by four x y coordinates in mm 58 picked in a counterclockwise if Y SCALE gt 0 or clockwise Gf Y SCALE lt 0 direction The best way to do this is to use the cursor to pick the points off of the display window The x y coordinates in mm are listed in the bottom row of the green format default example ALPHABETICAL KEYWORD DESCRIPTIONS box of the display circle defined by the x y coordinates of the center and the radius of the circle all in mm Drerangte defined by three x y coordinates Use the display program to pick these GNORE quadrilateral xl yl x2 y2 x3 y3 x4 y4 all values in mm cir
207. of disk I O error on rebooting Besides the window management files must have been corrupted because the PC icon styled login window disappeared instead it gives me a dumb terminal styled login window and every step runs very Slowly A check on usr adm SYSLOG found the error message as lt DENZO gt KILLED DUE TO INSUFFICIENT MEMORY SWAP By the way my machine has 32 Mb main memory It should be sufficient according to your manual My best uneducated guess is that the program s memory management might have some serious flaws Message killed is from operating system running out of swap not memory space Swap space is shared by all programs so it is dependent on how many processes are active you can run out of swap space in unexpected moments This has nothing to do which data set you process Rebooting the computer killed some other programs and increased the amount of swap space for Denzo however rebooting computers can make life difficult if your system is not well managed Find a better system manager HKLQ amp A We would like to use Denzo on our SUNs running Solaris 2 2 if that s not too much trouble for you They are 50 MHz 2 processor machines with plenty of memory so I d be surprised if they were inadequately slow For the time being the only other machine we could use for Denzo is our DECstation 5000 but that s slower Both SUN s and DECstation 5000 are OK but you run the risk of problems with the programs specific to these mach
208. of horizontal slow scanning direction pixels format RECORD LENGTH value_ pixels default FILM length RASTER Y SCALE this is defined by the FORMAT statement example RECORD LENGTH 2000 REFINE PARTIALITY REFINE NO PARTIALITY DO NOT USE format default Tells the program to determine the fraction partiality of the reflections on the image It is critical that this be on during the refinement of the crystal rotation parameters since CRYSTAL rotx and roty cannot be determined without knowing the partiality of the reflections on the image REFINE PARTI ALITY REFINE PARTI ALITY 67 THE HKL MANUAL Eb Do not use REFINE NO PARTIALITY see theoretical explanation in Refinement Using the Fit Command chapter REJECT Describes a condition whereby a reflection is rejected by Denzo This is especially useful for eliminating spurious measurements due to ice rings modifiers fraction The fraction of acceptable pixels in the background which are needed for the reflection to be accepted Default 0 75 i e 75 Very sensitive to small changes so increment by 1 2 if you are messing with this To get rid of spots in an ice ring increase this value by 1 2 until the preds in the ice ring are colored red After all the defining feature of an ice ring is a horrible background The acceptability of a background pixel is determined by the next three modifier words cutorr The o deviation from the best fit to the background plane
209. of judging the success of the experiment by the final statistics e g from Scalepack is not sufficient since it does not show if the experiment was done optimally The biggest problem with final statistics is that they do not differentiate well among the sources of problems and often come too late to fix them Therefore the experimenter must be aware of how the detector the X ray beam the crystal and the procedure all contribute to the final data quality and how each of them can make the experiment a failure Experimental Procedure In the traditional approach one collects data first and then starts analyzing the results This strategy has a risk that there may be a gross inefficiency in the setup of the experiment for example the data set may be incomplete the reflections may overlap the zones may overlap a large percentage of the reflections may be overloaded etc At that stage the only solution is to repeat the experiment which may be difficult with unique crystals or with experiments that require a synchrotron source Data collection is best performed as a highly interactive process Immediate data processing which the authors encourage provides fast feedback during data collection Most macromolecular crystallographic projects go through iterative stages of improving crystal quality and data collection strategy Typically most of the data collection time and effort is spent before the optimal point is reached Then if data collec
210. ofile fitting transformation of data to a reciprocal space coordinate system and the demonstration that a single oscillation image contains all of the information necessary to derive the diffraction intensities from that image The analysis and reduction of single crystal diffraction data consists of seven major steps These are 1 Visualization and preliminary analysis of the original unprocessed diffraction pattern 2 Indexing of the diffraction pattern 3 Refinement of the crystal and detector parameters 4 Integration of the diffraction maxima 5 Finding the relative scale factors between measurements 6 Precise refinement of crystal parameters using the entire data set 7 Merging and statistical analysis of the measurements related by space group symmetry The programs Denzo XdisplayF and Scalepack together form the HKL package THE HKL MANUAL Every January in Daresbury Laboratory the CCP4 Study Weekend 1S held Proceedings distributed by The Librarian Daresbury Laboratory Daresbury Warrington WA4 4AD UK contain comprehensive coverage of the X ray diffraction data analysis especially the volume Data collection and Processing January 29 30 1993 compiled by L Sawyer N Isaacs and S Bailey The HKL package can estimate Bragg intensities from single crystal diffraction data that are recorded on a position sensitive X ray also potentially neutron diffraction or electron diffraction detecto
211. ogram to merge combine average reflections with the same unique index This is the Scalepack default MERGE this is the default and need not be specified Allows you to input the value of the mosaicity of the data set This is normally read from the header of the x file MOSAICITY value read from the header of the x file MOSAI CITY 0 5 Opposite of the keyword ANOMALOUS Causes I and I to be merged This is the default NO ANOMALOUS 129 The HKL Manual default this is the default NO MERGE Flag for the output of unmerged reflections with the same unique Akl are not combined data Opposite of MERGE This is a very handy for specialized work This flag has subsidiary modifiers include partials original index or as default include no partials modifiers include partials include partials original index all observations both fully and partially recorded are included in the output The output will consist of the unique Akl batch number asymmetric unit I o and fractionality of the reflection There is no information about I and I although it may be possible to get this in subsequent versions of the program only fully recorded reflections and those fully recorded reflections created by the summation of partials are included in the output Partials which cannot be summed to a fully recorded reflection are lost The output will consist of the unique Akl batch number asymmetric unit I and o This i
212. oid later problems in data reduction Using the program for simulation of diffraction patterns also can be a tool for teaching crystallography 31 THE AKL MANUAL Refinement Using the Fit Command What to fit first FIT remains on FIT amp autoindexing 32 If autoindexing has given you a pretty good match between the observed diffraction pattern and your preds you can use Denzo to refine the parameters to several significant figures The crystal and detector orientation parameters require refinement for each processed image The refinement can be simple for a series of images collected with an on line detector or more complex if the detector orientation is only crudely known and varies from image to image as in the case of off line scanners The refinement is controlled by the user and can consist of several steps In each step the user defines the resolution limits and the order and number of parameters to be fitted Both detector and crystal parameters can be fitted simultaneously by the fast converging least squares method The refinement is done separately for each image to allow for the processing of data even when the crystal or the detector slips considerably during data collection Although Denzo command FIT ALL can successfully refine all the parameters crystal rotations film rotations camera distance unit cell dimensions etc simultaneously if their initial values are close enough in practice it is a good
213. ollection medium Identical to the keyword MAGI NG PLATE absorption length width rotation output file Total absorption of the film Default is 100 for image plates Distance from the top of the detector to the last useful pixel in the fast direction Depending on the detector on the computer display it could be the apparent width of the detector Width of the detector in the slow direction Depending on the detector on the computer display it could be the apparent height of the detector Describes the orientation of the spindle relative to the detector The value of the FILM rotation is function of which file storage convention see figure 5 in Detector Conventions analysis is being used For manual scanner adding 180 to film rotation can compensate for putting plate upside down into scanner Small changes refined by Denzo to film rotation can compensate for small rotation slop that is inherent when manually handling plates This gives the name of the processed output file FILM modifierl value output file absorption 100 length rotation format is a string of characters detector dependent detector dependent output file none FILM rotation 270 0 FILM absorption 2 ALPHABETICAL KEYWORD DESCRIPTIONS FIT FIX END FIT END FIX FIT and FI X able keywords format default examples These command words tell the program to FIT refine or FIX not refine a parameter specified
214. ommand file For example the modifiers rotz roty and rotx modify the keyword CRYSTAL as well as CASSETTE A precise description of each of these keywords and parameters can be found later in this manual 25 THE H KL MANUAL myexperiment dat 26 OSCILLATION range 1 0 OSCILLATION start 0 0 DISTANCE 150 0 SPACE GROUP P212121 use PI unless you know otherwise MOSAICITY 1 0 adjust to match spots TITLE my crystal SECTOR 1 to 90 RAW DATA FILE data mydir xtal0 osc data file note single quotes due FILM OUTPUT FILE data mydir xtalO x to special characters like X BEAM 99 4 Y BEAM 99 6 These are included only if they BOX 2 4 2 4 differ from those of the site default SPOT ELLIPTICAL 0 6 0 7 Otherwise you can omit them BACKGROUND elliptical 0 7 0 8 OVERLAP spot ERROR positional 0 050 PROFILE FITTING RADIUS 30 auto dat PEAK SEARCH FILE peaks file created by display program FIT x beam y beam cell crystal rotx roty rotz refine crystal and beam RESOLUTION LIMITS 100 4 1 use only if you want to exclude data from autoindexing Default is all data WRITE PREDI CTI ONS makes file hklpredictions GO When Denzo indexes an image it will also write a predictions file called hk lpredictions which contains the positions of the predicted reflections preds This file is read by the display program and the preds are superimposed on the diffraction pattern To do this go to the ImageWindow
215. on line detectors the range is from 1 to 2 1 65535 The particular OVERLOAD VALUE that is considered unreliable has been determined for several detectors and is input as the default when you specify the detector using the FORMAT keyword Denzo applies the overload table in Mar scanners so that the overload value should represent the highest reliable entry in the Mar overload table OVERLOAD VALUE integer units are function of detector type separate values for different IP formats contained in FORMAT defaults OVERLOAD VALUE 255 film E One of a kind detectors as well as the very old Mar scanners may have the wrong default value E The overload value is a data representation value before applying the data format dependent correction for strong intensities Eb Default values assume the full dynamic range of the instruments Mac Science Mar R Axis 63 THE HKL MANUAL PEAK SEARCH FILE peaks file Tells the program to perform autoindexing Get slightly confused PLATE ROTATION REDUNDANT This is exactly the same as the keyword FILM rotation Irecommend you stick with FI LM rotation to avoid getting confused with too many aliases POLARIZATION REDUNDANT Same as MONOCHROMATOR PRINT END PRINT 64 modifiers The PRI NT command tells the program to print the output specified by subsequent keywords to the log file It does not affect the results of the refinement Default values of the print sub options ca
216. on uniform exposure may be caused by spindle motor backlash shutter malfunction opening too early or too late 1onization chamber electronics failure if used decay or variation of the X ray beam intensity if ionization chamber is not used variable speed of the spindle motor etc Non uniform exposure is best diagnosed by exclusion of other problems that may affect data quality Graphical feedback provides confidence that the problem cannot be at the indexing integration stage INDEXING Absorption Large variations in absorption of X rays by the crystal will make data scale poorly and will produce visible variation of the background however it will not affect positional agreement The variation in the absorption can be avoided easily by a proper mounting of the crystal The correction for absorption is a whole field in itself Errors in detector orientation will produce distorted mapping from detector to reciprocal space A wrong crystal to detector distance will enlarge or decrease the apparent reciprocal space separation between Bragg peaks This error will not be uniform in three directions in the first approximation along the beam direction the error will be proportional to the square root of the error in the distance in the other two directions the error will have linear dependence In most cases a significantly wrong crystal to detector distance say an error of 10 will not make the autoindexing step fail immediately however the c
217. one and collected data cross a mirror plane 60 completion after 90 degrees data You can use option PRINT zones in Denzo Or you can redo the autoindexing with the desired values of SPINDLE AXIS and VERTICAL AXIS Also Scalepack can do the requested calculation Also is there some way to set a resolution cutoff for the peak search At 2A I have a dark nickel diffraction ring that peak search labels understandably with many peaks This hasn t yet skewed the auto indexing as there I can set RESOLUTION LIMITS but it would be nice to be able to avoid those areas in the data at the outset Specify RESOLUTION LIMITS 20 2 1 before autoindexing This will remove all spots found by peak search outside 20 2 1 A range If I have good cell parameters to start with can the indexing routine use them as given rather than recalculating them from the peak positions For one image that I have the Denzo autoindexing gives somewhat different cell parameters from what I think are the correct ones and the refinement then doesn t work as well as it should I d like to be able to try fixing the cell parameters and see if it helps type peak search file peaks file GO FIX cell It you did FIT cell in autoindexing UNIT CELL 98 99 115 90 90 90 and the program will use the UNIT CELL value Box definition Q CCP4 Sca Q ling HKLQ amp A you supplied This will work only if the autoindexing unit cell is close enough say within 1 2 of the un
218. pace group symmetry since the program at this stage of the calculation cannot distinguish lattice symmetry from pseudo symmetry The universal autoindexing procedure is now part of Denzo This procedure is able to deduce the lattice type crystal unit cell parameters and crystal orientation parameters from a single oscillation image It is also able to index more complicated lattices routinely for example body centered or rhombohedral lattices It really makes life a lot simpler Autoindexing in Denzo always finds a standard lattice however the crystallographer may prefer a non standard choice for example similar to one in a different space group Reindexing in Scalepack or manual indexing in Denzo accommodates such needs To start the autoindexing the first step is to select a set of spots from the displayed oscillation or still image A subset of these spots will be used by the autoindexing routine To select the spots go to the ImageWindow which contains the displayed image and click Peak Search The number of spots the peak search routine picks can be controlled with menu options which will appear after peak search finishes If the number of spots is too few or too many click the More Peaks or Fewer Peaks respectively as many times as necessary until you have about the right number of spots If your spots are unusually large or small for your detector choosing Size Peak Up or Size Peak Down will change peak search criteria peak s
219. perative Programming Workshops on Position Sensitive Detector Software at LURE Paris were dedicated to this task alone phase I and II May 16 June 7 86 phase HI November 12 19 86 Proceedings from these workshops contain the best but voluminous presentation of the theory The HKL suite implements most of the ideas discussed there Denzo amp XdisplayF Scalepack X ray data can be collected with 0 1 and 2 dimensional detectors O d single counter being the simplest and 2 d the most efficient in terms of measuring diffracted X rays in all directions Two dimensional detectors have been used from the very beginning of X ray diffraction studies the year 1912 Initially the 2 d detector was made of X ray sensitive film At present electronic and IP phospholuminescent best known by the trade name Image Plate detectors dominate To analyze single crystal diffraction data collected with these detectors several computer programs have been developed The 2 d detectors and related software are now used predominantly to measure and integrate diffraction from single crystals of biological macromolecules The usefulness of these systems in small molecule high resolution crystallography however is just being recognized and much of the rest of this discussion is applicable to this field as well The four most important recent developments in the data analysis of macromolecular diffraction measurements are autoindexing pr
220. pindle axis The VERTI CAL AXI S is the unit cell vector which is perpendicular to both the SPINDLE AXIS and the beam axis In practical terms this is simply the axis running in the vertical direction on the screen The way to determine each axis is to look at your diffraction pattern for two lines of spots running perpendicular to each other Call the one more parallel to the spindle axis the SPINDLE AXIS and measure the spacing between the spots using the mouse pointer If you are lucky you will have measured spacing corresponding to a major axis for example the c axis and you can enter the command SPINDLE AXIS O O 1 The same holds true for the VERTICAL AXIS But let s say you are unlucky or you have a centered space group and you can t see a major zone Here s what you can do L MANUAL INDEXING Page through your frames until you find one that does have a recognizable major zone Depending on your oscillation range this could mean looking at every 5 or 10 frame which is not such a bad investment of time If no major zone is apparent you will have to figure out which pseudo zone you are looking at For example let s say you have a tetragonal body centered unit cell with a b 120A c 130A You measure a spacing of about 90A This is clearly the 1 O 1 or O 1 1 axis There is another axis 90 degrees away from this one with spacing 62A This must be the 0 2 0 or 200 axis and can be entered for the vertical axis Once the a
221. r for example film IP scanners or charge coupled device CCD area detectors The programs allow for data collection by oscillation Weissenberg and precession methods The detector can be either flat or cylindrical The detector readout can be either rectilinear or spiral although spiral coordinates must be converted to rectilinear before processing The package allows for random changes in the position and the sensitivity of the detector between consecutive exposures The package features profile fitting weighted refinement eigenvalue filtering and a universal definition of detector geometry Denzo is a program which takes raw X ray diffraction data for example from a scanned piece of film or imaging plate and reduces it to a file containing the h k indexes and background and L P corrected intensity of the spots on the image along with an estimate of the error In this way the useful information is extracted from the image Depending on the number of spots on the image a typical 7 to 18 MB imaging plate raw data file or in case of the new RAXIS IV scanner and 50 micron scan the size of the image is 72 MB will yield a reduced file of approximately 0 1 to 2 0 MB Advantages of using Denzo to process raw X ray diffraction images The use of a flexible weighted profile fitting algorithm for measuring the intensity of the spots provides highly accurate data processing Crystal and detector parameters can be refined in any ord
222. r real space vectors is completed the program finds the three linearly independent vectors with minimal determinant unit cell volume that would index all or more precisely almost all of the observed peaks These three vectors form a basis but are unlikely to form a standard basis for a description of the unit cell The process of converting a basis into a standard basis is called cell reduction The program follows the definitions in the International Tables for Crystallography and finds the best cells for all of the fourteen Bravais lattices The transformation of the primitive cell to a higher symmetry cell may require some distortion of the best triclinic lattice that fits the peak search list Due to experimental errors the fit is never perfect for the correct crystal lattice Sometimes the observed reflections can fit a higher symmetry lattice than one defined by space group symmetry Such a condition is called lattice or metric tensor pseudo symmetry If this happens the lattice determination and assignment of lattice symmetry may get complicated The procedure in such case is to index the data in lowest symmetry lattice that does not introduce wrong lattice symmetry triclinic lattice is always a safe choice and look for the symmetry of the intensity pattern during the scaling of symmetry related reflections Denzo calculates the distortion index for all fourteen of the Bravais lattices It is up to the user to define the lattice and s
223. ractor that we would like to get some cell constants on and in particular determine whether it is monoclinic or orthorhombic One image does not seem to be enough to help Denzo make a clean distinction Sorry it is impossible It may be possible in the future but it will require substantial change to the program code You can make clean distinction of monoclinic from orthorhombic during scaling 143 The HKL Manual 144 I am trying to process data collected on an R axis using Denzo and my problem is the following Although I have what seems to be a pretty decent orientation using the R axis software when I import an oscillation frame into Denzo the overlap between predicted and observed spots is far from decent Some one suggested that I try to obtain a conversion between the angles Q p as defined in R axis and those used by Denzo There is conversion for the angles for some space groups however it is much simpler to do autoindexing in Denzo If you still have a problem it is most likely due to misindexing Denzo reverses the definition of x and y axis relative to R axis software If you enter X BEAM and Y BEAM swapped Denzo may misindex Before we indexed this crystal on an orthorhombic lattice with cell constants of 133 7 67 6 35 7 Now the unit cell is primitive monoclinic 61 89 103 76 70 37 90 109 27 90 and distortion index for orthorhombic is 7 36 This is why Denzo always outputs 14 best fitting Bravais lattices when a
224. rbitrary units Partiality fraction of the intensity of the reflection extected in this image Diffraction angle degrees The last may be very useful when preparing input to program SCALA Is the Denzo output file format dependent on the type of detector It is now the same total number of reflections output from Denzo 174715 unique reflections 3873 reflections in reject dat file 1398 output from Scalepack 73161 From this I think I ve only rejected about 2 of data No only 1398 174715 about 0 6 Below is a bit of a x file from Denzo Notice the middle reflection The profile fitted intensity and theo are not fn 1 but seem to be integer data Is this an intended part of the program to deal with overflows or have I run into an operating system bug Do you think this will only happen on reflections flagged with a negative o 2 16 28 1 134 8 191 2 1 37 25 3 0 995 1371 5 1284 9 0 191 3252 4 O 12 32 1 118141 5399 24 77 10031 0 997 1372 8 1196 3 0 191 3094 7 o 36 O T O09574 95300 one SIS ak 0996 Lehane HOOD URLIA 509 Yes this an intended part of the program to deal with overflows The problem is that the Denzo x file does not have information about the angle of the individual reflection I know it has information about the starting ending 9 of the batch but NOT the o of each and every reflection which is the ROT column used in the CCP4 package This is the problem although rotavata and agrovata do not need this in
225. rd 29 68 71 143 RESOLUTI ON Scalepack keyword 136 resolution shells 90 RESOLUTI ON STEP Scalepack keyword 136 Reverse XdisplayF button 19 Rhombohedral 158 164 R linear 135 Rmerge unweighted error measure 8 10 88 rotaprep 146 R squared 135 S saturated image pixels 19 SCALA 146 SCALE ANOMALOUS Scalepack keyword 91 136 SCALE RESTRAIN Scalepack keyword 137 SCALEPACK 42 scaletwosca com command file 101 scattering 68 Scenarios 94 screen output 26 screening through your frames 19 SDMS Hamlin gt detectors 127 sector 157 SECTOR Denzo keyword 69 70 SECTOR Scalepack keyword 137 SECTOR WIDTH Scalepack keyword 137 separators 46 SGI 147 Show overflow XdisplayF button 19 shutter 96 151 shutter malfunction 8 sigma cutoff 90 137 SI GMA CUTOFF Scalepack keyword 90 137 simulation as a teaching tool 31 simulation of diffraction pattern 8 single measurements 90 site dat command file 13 example 25 SKEW Denzo keyword 33 69 slope 68 small molecules 5 10 29 52 88 software agrovata 150 rotavata 150 Scala 150 SPACE GROUP Denzo keyword 69 as replacement to LATTI CE keyword 60 Space Group Identification 117 SPACE GROUP Scalepack keyword 138 space groups 8 spindle axis 78 SPINDLE AXIS Denzo keyword 54 70 78 SPINDLE AXIS Scalepack keyword 131 139 spindle motor backlash 8 Spindle beam orinetation system 48 SPOT 36 63 distortion 55 elliptical 36 radius
226. rea is flagged by a negative sign applied to the o estimate Profile fitted intensities seem to be reliable independent of the existence of such pixels in the spot area Ice rings are sometimes seen in diffraction images from frozen crystals but they need not ruin a data set There are two places in the indexing process where ice rings can cause problems The first is during autoindexing and the second is during the refinement and integration steps after getting the initial orientations Ice rings in autoindexing The first thing to do if you have ice rings is to simply ignore them and run autoindexing as usual Denzo edits the peak list produced by the Peak Search routine of XdisplayF and may succeed in autoindexing the image regardless of whether the display program picked a lot of 39 THE H KL MANUAL Ice rings in your data ice non protein lattice peaks You can tell if this worked by inspecting the image after autoindexing and then again after running the refine macro If the preds line up on real spots then autoindexing worked despite the ice ring If the ice ring does interfere with autoindexing here is a simple remedy adjust your resolution limits to exclude the ice ring This means you would edit the auto dat macro to include the resolution limits In the example provided above the resolution limits were commented out You could exclude the ice ring from autoindexing by setting the high resolution limit to about 4 1 A Once
227. s It will reduce the output intensities by the factor entered format DEFAULT SCALE value default 1 example DEFAULT SCALE 10 E Useful to reduce overall scale of the data set If the numbers in the output file are too large DEFAULT SCALE 10 will reduce them 10 fold 122 THE SCALEPACK KEYWORD DESCRIPTIONS DO NOT REJECT OUTLIERS Turns off the reject outliers flag format DO NOT REJECT OUTLIERS default DO NOT RE ECT OUTLIERS outliers not rejected automatically but see REJ ECT OUTLIERS for more discussion END FIT Provides a logical end for the fit command At the moment this has no effect but may be implemented in future versions of the program ERROR SCALE FACTOR This is a single multiplicative factor which is applied to the input or This should be adjusted so the normal y goodness of fit value that is printed in the final table of the output comes close to 1 By default the input errors are used ERROR SCALE FACTOR 1 It applies to the data which are read after this keyword so you can apply different error scale factor to subsequent batches by repeating this input with different values format ERROR SCALE FACTOR value default ERROR SCALE FACTOR 1 example ERROR SCALE FACTOR 1 3 good starting value for format denzo_ip ESTIMATED ERROR Estimate of the systematic error for each of the resolution shells There must be exactly the same number of error estimates here as there are NUMBER OF ZONES So if you have 10 zones
228. s and the refinement proceeds normally amp 1 THE HKL MANUAL Detector Specifications Detector Description Modifier XdisplayF Ver 1 5 and lower fuji 10 bit scanner 100 micron raster Film scanner Purdue type 25 micron scan Film scanner Purdue type 50 micron scan Film scanner Purdue type 100 micron scan Film scanner Uppsala type 50 micron scan Film scanner Weizmann type 50 micron scan CCD Bruker quad detector IMCA beam line CCD Bruker quad detector binned mode IMCA beam line CCD SMART 6000 Bruker detector CCD Bruker single detector older CCD SMART 6000 Bruker detector in binned mode CCD APS 1 detector at SBC CCD APS 1 detector at SBC in binned mode CCD SBC 2 detector at SBC CCD SBC 2 detector at SBC in binned mode Nonius KappaCCD detector Nonius KappaCCD detector dezingering on Nonius KappaCCD detector binned Nonius KappaCCD detector binned dezingering on Nonius KappaCCD 2000 detector Nonius KappaCCD 2000 detector dezingering on Nonius KappaCCD 2000 detector binned Nonius KappaCCD 2000 detector binned dezingering on CCD Brandeis University at BNL X12C 82 INST_FUJI INST_FILM25 INST_FILM50 INST_FILM100 Modifier XdisplayF Ver 1 96 and higher fuji 10bit film 25 film 50 film 100 film lars film is ccd bruker quad ccd bruker quad binned ccd smart6000 ccd bruker ccd smart6000 binned ccd aps1 ccd aps1 binned ccd sbc2 ccd sbc2
229. s and when computing statistics but are merged separately and output as I and I for each reflection ANOMALOUS not turned on I and I are combined ANOMALOUS 121 The HKL Manual B RESTRAIN Can be used to restrain B factor differences from consecutive films or batches The value which follows the flag represents the amount in A you will allow the B factors to differ from consecutive frames or batches See also the keyword SCALE RESTRAIN Format B RESTRAIN value_ A default not turned on example B RESTRAIN 0 5 BACKGROUND CORRECTION The BACKGROUND CORRECTION command specifies amount per frame to increase background Corrects for errors in SDMS Hamlin data integration in the software distributed in the late 1980 s and early 1990 s The value given after the keyword increases the background by counts frame Valid for archive file format only format BACKGROUND CORRECTION value_ counts frame default not turned on example BACKGROUND CORRECTION 15 DEFAULT B FACTOR Overall B used only in the absence of an I NI TIAL B FACTOR You can apply a higher value to lower your Rmerge Does not affect the quality of the data gt Applies only to data read in after this command format DEFAULT B FACTOR value default O example DEFAULT B FACTOR 5 DEFAULT SCALE Overall scale factor used in the absence of an initial scale factor This is useful if the data are too strong which is sometimes the case with small molecule
230. s because the xengen or XDS MADNES or Scalepack software has already done these tasks You would perform the iterative refinement with this data just as described above for Scenario 1 Eb Some of the programs produce binary files that Scalepack may not be able to read in if Scalepack runs on a different computer architecture than data reduction program that produced binary file 97 The HKL Manual Scenario 3 Heavy Atom Search Scale native plus a few frames of a potential derivative scalepotderiv com The Native Dataset NUMBER OF ZONES 10 ESTIMATED ERROR 0 0 0 0 0 0 0 0 0 0 OAC OOS OIRO 10 510 10 210 ERROR SCALE FACTOR 1 0 SPACE GROUP P3121 FORMAT Scalepack FILE 101 native sca REFERENCE BATCH 101 Potential Derivative RESOLUTION 3 5 ESTIMATED ERROR 0 05 0 05 0 05 0 05 O 05 0 05 0405 0205 U0 0405 ERROR SCALE FACTOR 1 6 REJECTION PROBABILITY 0 0001 GNORE OVERLOADS ADD PARTIALS 1 to 3 FORMAT denzo_ip HKL MATRIX 1 0 0 SECTOR 1 to 3 FILE 1 NewHo x OUTPUT FILE Jjunk sca number of resolution shells for statistics ESTIMATED ERROR for each resolution shell Set to zero because native has already been scaled don t need here because data already scaled SPACE GROUP of native amp derivative duh native data has been processed already and is in Scalepack format REFERENCE BATCH in scaling procedure We will compare derivative to native so native is ref
231. s nl weis nl format Weisenberg exp Photon factory fuji BA 100 scanner SL INST WEIS weis sl weis sl format Weisenberg exp 85 THE HKL MANUAL Detector Description Modifier XdisplayF Modifier XdisplayF Ver 1 5 and lower Ver 1 96 and higher Photon Factory beam line 18b whole pf beam line 18b image Photon Factory beam line 18b center of beam line 18b center Photon Factory beam line 18b beam pf beam line 18b beam region Photon Factory beam line 18b bottom pf beam line 18b bottom Photon Factory beam line 18b top pf beam line 18b top 86 Denzo Format Modifier pf beam line 18b pf beam line 18b pf beam line 18b pf beam line 18b pf beam line 18b ABOUT THE SCALEPACK MANUAL THE SCALEPA CK Manua About the Scalepack manual from Edition 4 The manual is organized somewhat differently than the Denzo manual Compared to film or IP processing scaling is complex There s a lot more to scaling data than just plugging in a bunch of Denzo x files and getting a merged data set out the other side In particular scaling is where you make explicit estimates of the errors associated with your experiment and these error estimates are crucial in determining the statistics by which you judge the quality of your data In addition Scalepack is also used for many other manipulations of the diffraction data As a result we decided to make the heart of the manual a series of scenarios each accompanied by a sample command file whi
232. s overwritten You can specify the threshold for what you consider to be a rejectable probablity The default is 0 9 which is fairly safe but you may want to decrease this to say 0 5 on later rounds of rejection format WRITE REJECTION FILE value default does not write the rejection file but if it does the default value is 0 9 example WRITE REJECTION FILE 0 5 140 HKLQ amp A HKLQ amp A This chapter presents a selection of the questions sent to the Akl server over the past year or two They were selected to illustrate points which may not be covered in the main portion of the Denzo and Scalepack manuals Answers are by Zbyszek Otwinowski and Wladek Minor Spelling and grammar have been cleaned up only to a limited extent The general topic headings are listed here 20 Absorption correction factors Auto Editing Autoindexing Box Definition CCP4 Scaling Computer problems and questions Confusion amp other Consequences of Profile fitting errors in Denzo Corrections Denzo Output Denzo Problems Display Example of Bad Detector Pathology Ignore Mosaicity Negative Intensities Overload Refinement Rhombohedral Lattices Twinned Crystals 141 The HKL Manual 20 Q In HKL manual we found a notation that 20 could go into CASSETTE rotx However this doesn t seem to work According to the R axis we have set 20 to 10 degrees Does the 20 option work correctly The 20 swing is described by CASSETTE rot x
233. s the default for NO MERGE There is no information about I and I the output will also contain the original not unique Akl for each reflection This is designed for MAD local scaling work The original index modifier only works with the default NCLUDE NO PARTIALS The output will consist of the original Akl unique hkl batch number a flag 0 centric 1 I 2 I another flag 0 hkl reflecting above the spindle 1 Akl reflecting below the spindle the asymmetric unit of the reflection I scaled Lorentz and Polarization corrected and the o of I The format is 614 16 212 13 28 1 format NO MERGE modifier example NO MERGE original index This command is also useful if you just want to combine all of the information contained in multiple x files into a single file Simply read in all of the x files don t do any scale or B factor refinement and output NO MERGE include partials NO PROFILE TEST This is a flag which operates only on ARCHIVE format files Tells Scalepack that reflections with weird profiles should not be rejected Opposite of PROFILE TEST format NO PROFILE TEST default this is not the default NUMBER OF ITERATIONS 130 Number of cycles for refinement of scale and B factors Default value is 20 cycles If it is set to 0 the program computes statistics merges reflections with the same unique Akl and writes the output file based on the initial scale and B factors Normally you do no
234. s to make the polarization positive For a laboratory source use modifiers graphite if graphite monochromator is installed or filter if X ray beam is filtered or mirrors are installed modifiers filter No polarization value 0 graphite Reflecting plane perpendicular to the plane formed by the spindle and the beam CuKa and MoKa radiation Value of 0 11 for CuKa or 0 03 for MoKa format MONOCHROMATOR modifier default 0 0 example MONOCHROMATOR graphite POLARI ZATION 0 11 E The old keyworded values for the synchrotrons are probably wrong now Always ask the synchrotron staff what the correct number is A good guess for a bending magnet source is polarization 0 9 For a wiggler source it can vary a lot from station to station in the range of 0 0 to 0 9 MOSAICITY Defined in Denzo as the rocking angle in degrees in both the vertical and the horizontal directions which would generate all the spots seen on a still diffraction photograph It includes contributions due to X ray bandwidth beam crossfire etc format MOSAICITY value degrees default this is a required input example MOSAICITY 0 5 61 THE HKL MANUAL NO BEAM format default Tells the program not to use the beam spot in the refinement If the option USE BEAM has been invoked earlier in the refinement it should be revoked with the NO BEAM keyword in the later Stages NO BEAM this is the default option see USE BEAM NUMBER OF RECORDS
235. s will have the same number of reflections Changing the resolution step could be useful to prepare a table of statistics to compare with other programs which may print out the statistics differently format RESOLUTION STEP value default RESOLUTION STEP 3 example RESOLUTION STEP 2 5 SCALE ANOMALOUS This is the flag for keeping Bijvoets I and I separate both in scaling and in the output file If the SCALE ANOMALOUS flag is on anomalous pairs are considered non equivalent when calculating scale and B factors and when computing statistics and are merged separately and output as I and I for each reflection E This is a dangerous option because scaling may be unstable due to the reduced number of intersections between images The danger is much larger in low symmetry space groups SCALE ANOMALOUS will always reduce R merge even in the absence of an anomalous signal because of the reduced redundancy However y s will not be affected in the absence of an anomalous signal format SCALE ANOMALOUS 136 default THE SCALEPACK KEYWORD DESCRIPTIONS definitely not a default SCALE RESTRAIN format default example SECTOR format default example SECTOR WIDTH format default example SIGMA CUTOFF E 1 2 Can be used to restrain scale factor differences from consecutive films or batches The value which follows the flag represents the amount you will allow the scale factors to differ from consecutive films
236. sage like insufficient space in common spots losing reflection What does that mean Denzo is running out of memory and ignores some reflections This may happen if the BOX is unreasonably large Reduce the BOX size or use the DenzoBig version of the program 154 HKLQ amp A Would you make a larger version of Scalepack as I have some 2 8A data on an 1222 form of a virus crystal a 328 2 b 341 8 c 363 75 and Scalepack can t quite cope with the number of reflections Use ScalepackBig or Scalepack VeryBig We collect our data on an Raxis VAX workstation and transfer it using FTP to the SGI We have no trouble getting the Denzo to run on the oscillations but the display program the one we got is INST CORNELL gives a message that the data file may be corrupted The sample data that was packed in the tar file runs just fine Is this the right version of the display program Use INST_RX instead INST_CORNELL is design to read in data scanned with KODAK scanner at Cornell now decommissioned I believe that my copy of the Scalepack executable file is somehow corrupted because every time I run it it always return with strange error message CRYSTAL6 SCALEPACK SCALEPACK SYNTAX ERROR AT LINE 1 A_ H gt Q CE FMH P B A K A Yes you are correct This is not the right executable Scalepack runs it reads the frames but then just sits there I used the example as a guideline and I also tried with deleting the postrefinem
237. sionally you may try to scale data which was collected in very thin sectors or in sectors whose width is less than the mosaicity or from data which consists of only a few frames In these cases there may not be enough intersections between frames to get accurate scale and B factors Indeed what you may see is both scale and B factors ranging all over the place If things get really bad the program may crash due to floating point arithmetic exceptions when taking the exponent of unreasonable B factors This does not mean that the data is unusable it simply means that the scale and B factors must be restrained The keywords SCALE RESTRAIN and B RESTRAIN are used for this purpose and an example of there use is shown below Note that the restraints only apply to the frames over which you are adding up partially recorded reflections so you must have the ADD PARTIALS keyword in the command file for this to make any sense 10 number of resolution shells for statistics 0 04 0 04 0 04 0 04 ESTIMATED ERROR for each resolution shell 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 3 multiplicative factor applied to input o REJECTION PROBABILITY 0 0001 expected fraction of outliers in the data reject read in rejected reflections WRITE REJECTION FILE 0 9 obs with greater than 90 chance of being outliers are written to the SPACE GROUP P21 reject file OUTPUT FILE mydata sca OUTPUT FILE containing h k 1 I REFERENCE BATCH 1 REFERENC
238. spot and box sizes it s time to process the rest of your frames This can be done as a batch job or as a hands off interactive session Wladek and Zbyszek recommend that you take half an hour and watch your frames refine and integrate interactively because this is the best way to spot any problems For example if you have misindexed the first frame by one due to an incorrect beam position you are likely to see it very quickly this way Since you still have your DenzoWindow and ImageWindow open just continue on from there using the following refinement macro FIX all good to do just in case START REFI NEMENT beginning of refinement loop PRINT no shifts no profiles limits output insert beam search routine here if necessary see below RESOLUTION LIMITS 100 2 8 don t use highest res yet FIT crystal rotx roty rotz refine crystal orientations GO GO GO RESOLUTION LIMITS 100 2 2 highest resolution FIT ALL fit other parameters GO GO GO GO GO PRINT profiles 1 1 print average profile in single frame CALCULATE GO end of refinement loop This macro will refine all of the subsequent frames specified by the SECTOR in the macro myexperiment dat p 26 You can just sit back and watch them zip along on the screen The result will be a series of x files called xtalo x Alternatively you can process everything in batch mode using the following command file denzo com setenv cnverr yes this line is only for IBM RS 6000 cd people my
239. stem It also needs a substantial amount e g 1 5 GB of swap space Note that if you are indexing 72 MB Raxis IV images you will need 144 MB of memory In order to run XdisplayF open a window on a graphics terminal ImageWindow and launch the program Launching the program involves the following calling the program itself xdisp is an alias calling the format variable the same as the modifier of the keyword FORMAT in Denzo telling the program the name of the image file to be displayed Here s an example xdisp raxis xtal osc 1 In this example the name of the frame to be displayed is xtal001 osc as number represents the sector frame number The old versions of the program used a little different convention xdisp raxis xtal001l osc hklpredictions Communication between Denzo and XdisplayF is achieved by using hklpredictions and test files This explains why the programs must be started in the same directory The program version number is stored in these files so you should delete them when switching to a new 17 THE H KL MANUAL version of the programs hk precictions is a name of the file which Denzo creates It contains the predicted reflections The use of hklpredictions file is a default therefore is put in parentheses The old convention still works for some detectors The change of convention has been forced by the variety of naming schemes used by detector vendors and developers The new convention has
240. stions addressed to the program Authors About the Manual from Edition 4 This manual began life as a simple compilation of some of the more useful things I D G needed to remember when indexing X ray diffraction images with Denzo This was especially important with earlier versions of the program which lacked both the autoindexing option and a sophisticated graphical display package For various reasons Zbyszek convinced me to expand these initial efforts into a comprehensive manual the fruits of which you see here The emphasis of this manual is on how to run Denzo and Scalepack for typical applications There is a keyword documentation section later in the manual for those who wish to learn more about the features of the program Every effort has been made to make this manual as up to date and as accurate as possible Since the program continues to undergo constant revision and development however it cannot be promised that the version of the HKL package you are using is exactly the same one described here Fortunately most of the new features of the HKL package do not preclude the older commands so it is usually possible to continue to use the program without taking advantage of the new stuff As I originally conceived it this manual was never meant to be a polished boring professional product and occasionally you may notice that I use slang and make various insinuations which could but really should not be interpreted as
241. surement uncertainties By default the Rmerge statistic in Scalepack does not reject observations In contrast most other programs define the Rmerge statistic with a large number of observations deleted from Rmerge This difference in the definition of Rmerge a more detailed discussion is in the Scalepack program description between programs frequently is very large Quoting Rmerge Statistics without referring to the program definition used does not properly describe the crystallographic experiment Until 1988 Denzo and Scalepack were developed at the University of Chicago from 1989 through 1994 at Yale University and currently at the University of Texas Southwestern Medical Center at Dallas XdisplayF was developed at Purdue University until 1995 and currently at the University of Virginia in Charlottesville We would like to thank Janet Smith Anne Bloomer Alan Friedman John Tesmer Iwona Minor Bob Sweet for comments and contribution to programs and documentation Michael Sintchak for the first UNIX port of programs Denzo and Scalepack Paul Sigler for his encouragement in this project and Michael Rossmann for providing the initial stimulus and ideas for the first implementation of the programs We also would like to acknowledge Janet Smith for the initial version of the Scalepack manual INTRODUCTION Introduction The theory behind data reduction methods 1S complex enough that a series of EEC European Economic Community Coo
242. t hardware recommendations but note that this is not meant to be a complete list of computers on which the program runs It s more to give you an idea of the direction your crystallographic computing should be heading DEC Alpha SGI Pentium IILIV Operating System OSF UNIX Linux Display 17 21 1280x1024 or more 16 bit or more workstation not X terminal Disk gt much more that one data set local to workstation gt preferably more than 20 GB Swap Space gt 1 5 Gb very important gt You should never economize on swap space because otherwise the programs may get killed randomly by the operating system m OSF Digital Unix and IBM AIX set default datasize to unrealistically low values Datasize should be set to unlimited when the HKL programs are run 77 THE HKL MANUAL Manual Indexing of Diffraction Images Manually indexing and orienting diffraction patterns is a tricky business more so with small unit cells centered space groups or lattices containing pseudosymmetry and less so with high resolution data and simple lattices Suffice it to say that the more you know about your crystal and detector parameters the easier the job will be For the purposes of this section I will assume that you know the following e detector type duh e wavelength e crystal to film distance within 1 or so e x beam and y beam within a pixel or two each take a quickie direct beam shot if unsure e detector 20 angle with
243. t dat auto dat begins autoindexing Step 6 ImageWindow Make sure green circles corresponding to whole reflections cover see Getting that First real reflections Partial reflections are displayed as yellow circles If Orientation Indexing circles don t cover real spots sorry gotta read more manual DenzoWindow Examine the distortion table Note the highest symmetry lattice with low distortion Hopefully this corresponds to the lattice type you entered in the file myexperiment dat If not exit both windows edit myexperiment dat appropriately and repeat from Step 4 in the Denzo Window Step 7 ImageWindow Hit Zoom wind Select a region of your pattern to examine Hit the Int Box button Now you are going to make sure your integration box parameters are correctly set This exploits the advantage of 13 THE HKL MANUAL see Mosaicity and Spot Shape section in Refinement Using the Fit Command DenzoWindow ImageWindow DenzoWindow ImageWindow DenzoWindow ImageWindow DenzoWindow Denzo interactive analysis Examine the diffraction pattern to determine the resolution limits The resolution at the position of the cursor is displayed in the green box on the right side of the window Let s say it s 2 2 A RESOLUTION LIMITS 100 2 2 FIT ALL GO GO GO GO GO Examine the diffraction pattern Are all the real reflections covered by predicted reflections preds If not you will have to increase the
244. t have to specify this unless you want to avoid scaling THE SCALEPACK KEYWORD DESCRIPTIONS format NUMBER OF ITERATIONS integer default 20 iterations example NUMBER OF ITERATIONS O NUMBER OF ZONES Number of resolution shells the data is divided into for the basis of calculating statistics This input is required and must match the number of zones specified under the ESTI MATED ERROR keyword Handy tip it s nice to set up the NUMBER OF ZONES to equal the number of zones used by X p or for the output of refinement statistics by shell for when you get around to publishing your data and refinement statistics together in a paper format NUMBER OF ZONES integer default this input is required example NUMBER OF ZONES 10 ORIENTATION AXIS 1 Three integer vector which describes the orientation of the vertical axis of the crystal Equivalent to Denzo and Scalepack keywords VERTI CAL AXI S This information is usually not input by the user but subsequent versions will read it from the header of the x files This command does not affect scaling or postrefinement however it does affect the values of the misorientation angles reported in the Scalepack log file format ORIENTATION AXIS l integer integer integer default 1 0 0 example ORIENTATION AXIS 1 0 1 O ORIENTATION AXIS 2 Same as ORIENTATION AXIS 1 except describes the spindle axis of the crystal Equivalent to Denzo and Scalepack keywords SPI NDLE AXI S Otherwise the same as abov
245. t to zero because native has already been scaled Multiplicative factor applied to input o s_ expected fraction of outliers in the data 101 The HKL Manual Scenario 5 Reindexing Reindexing involves reassigning indices from one unit cell axis to another This becomes an important issue when comparing two or more data set which were collected and processed independently This is because Denzo when confronted with a choice of more than one possible indexing convention makes a random choice This is no problem except that if it makes a different choice for a second data set the two will not be comparable without reindexing one or the other One cannot distinguish non equivalent alternatives without scaling the data which is why this is not done in Denzo You can tell if you need to reindex a data set if the y values upon merging the two are very high e g 50 This makes sense when you consider that scaling two or more data sets involves comparing reflections with the same hkl or index If the two indexing schemes are equivalent but not identical chaos will result No reindexing no new autoindexing and nothing except changing the sign of Y SCALE in Denzo can change the sign of the anomalous signal Below are nine reindexing scenarios which should cover all of the possibilities encountered in protein crystallography Switch a and cin P21 or P2 102 NUMBER OF ZONES 10 number of resolution shells for statistics ESTIM
246. ta Not surprising Ask Eleanor Dodson author of Rotaprep or Phil Evans author of SCALA Computer problems and questions Q 146 T I have problems with Denzo package Here we have the Open VMS AXP v 1 0 Running any exe I get this message run ccd S DCL W ACTIMAGE error activating image LIBOTS CLI E IMGNAME image file DKA300 SYSO SYSCOMMON SYSLIB LIBOTS EXE SYSTEM F SHRIDMISMAT ident mismatch with shareable image Thinking that this error was due to fact that I have an old version I got the LIBOTS from the v 1 5 I still have the same error You need Open vms AXP v 6 1 to run Denzo Upgrade your operating system Is the new Scalepack slower than the old one Scalepack speed is about the same Postrefinement may take a lot of extra time and very old versions had no postrefinement however postrefinement is optional now so it can be as fast as the old one Subjective variation in the speed of the program may be strongly affected by other programs on the same computer in particular if the operating system starts to swap programs While I was trying to index the data from a twinned crystal the Denzo always quit on me with a simple error message Killed Then I realized that it occured to me once when I was processing the other data set from a good crystal At that instance I logged off and rebooted the system Afterwards everything was OK However this time the system is corrupted seriously with error message
247. ta without first making a sophisticated judgment about which data is good and which is bad To accurately describe the structure of a protein molecule we ordinarily need higher resolution data than the crystal provides That is life One of the main judgment the crystallographer makes in assessing the quality of his data is thus the resolution to which his crystal diffracts In making this judgment we wish to use the statistical criteria which are most discriminatory and which are the least subjective In practice there are two ways of assessing the high resolution limit of diffraction The first is the ratio of the intensity to the error of the intensity i e I o The second way which is traditional but inferior is the agreement between symmetry related reflections 1 Rinerge 89 THE HKL MANUAL 90 From a statistical point of view I o is a superior criterion for two reasons First it defines a 9 resolution limit since by definition I o is the signal to noise of your measurements In contrast Rmerse 1S not directly related to signal to noise 8 Second the o assigned to each intensity derives its validity from the X s which represent the weighted ratio of the difference between the observed and average value of I I squared divided by the square of the error model the whole thing times a factor correcting for the correlation between I and I Since it depends on an explicit declaration of the expected error in
248. th a single 3 degree oscillation 99 The HKL Manual Scenario 4 Scale Multiple Native Data Sets Together Data from Denzo 100 There are two ways to scale multiple native data sets together The first and preferred way is to scale all of the x files together If you do it this way you get to do postrefinement over all of the raw data and the scaling is applied at the level of the individual frames rather than over the whole data set This 1s probably a little more accurate Also it saves you from having to run Scalepack twice once for the second native data set and then to scale the two sca files together scaletwo cam NUMBER OF ZONES 10 ESTIMATED ERROR 0 04 0 04 0 04 0 04 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 3 REJECTION PROBABILITY 0 0001 WRITE REJECTION FILE 0 9 SPACE GROUP P21 REFERENCE BATCH 1 reject POSTREFINE 10 FIT crystal a 1 to 327 FIT crystal b 1 to 327 FIT crystal c 1 to 327 FIT crystal beta 1 to 327 FIT baten rotx 1 to 327 FIT batch roty 1 to 327 FIT batch mosaicity 1 to 327 ADD PARTIALS O to 103 201 to 327 FORMAT denzo_ip per erate Set i x SECTOR 1 to 103 FILE 1 my_crystall x Data Set 2 x x SECTOR 1 to 127 FILE 201 my_crystal2 x OUTPUT FILE my_crystall2 sca number of resolution shells for statistics ESTI MATED ERROR for each resolution shell Multiplicative factor applied to input os expected fra
249. the SECTOR argument to be incremented by 1 Defines the orientation and the shape of the detector historically a film cassette rotz the three angular deviations from zero LOUY rotx radius radius of the film in a curved cassette m Radius of curvature does not have to be equal to the film to crystal distance and may be different than the radius of the cassette due to the thickness of the film or imaging plate flat turns off radius CASSETTE rotx valuel roty value2 rotz value3_ degrees CASSETTE radius valuel_ mm all angles 0 0 CASSETTE rotx 2 0 roty 140 You rarely have to enter these values since the program can refine them once the crystal 52 parameters are established However some computer time can be saved if previously ALPHABETICAL KEYWORD DESCRIPTIONS determined values are used Usually they change significantly only after the camera is realigned There is no limit on values E The cassette rotations are the primary means by which a non zero 20 angle is specified CELL COUPLING format default example CROSSFIRE modifiers format defaults example along with DISTANCE X BEAM and Y BEAM For example a 20 setting of say 20 degrees R Axis convention would be described by CASSETTE rotx 20 0 note change in sign The DI STANCE will have to be changed as well since it is the distance from the crystal to the point where it intersects the IP surface This can be calculated by the
250. the measurement the user of the program is part of the Bayesian reasoning process behind the error estimation The essence of Bayesian reasoning in Scalepack is that you bring X or technically speaking the goodness of fit which is related to the total X by a constant close to 1 0 by manipulating the parameters of the error model Ryerge on the other hand is an unweighted statistic which is independent of the error model It is sensitive to both intentional and unintentional manipulation of the data used to calculate it and may not correlate with the quality of the data An example of this is seen when collecting more and more data from the same crystal As the redundancy goes up the final averaged data quality definitely improves yet the Rose also goes up As a result Rmere 1s only really useful when comparing data which has been accumulated and treated the same This will be discussed again later In short I o is the preferred way of assessing the quality of diffraction data because it derives its validity from the X likelihood analysis Unless all of the explicit and implicit assumptions which have been made in the calculation of an Rmerge are known this criterion is less meaningful This is particularly true when searching for a number which can be used by others to critically evaluate your work There are two modes of analysis of data using X s The first mode keeps the X or more precisely the goodness of fit const
251. the point is that errors of this type can persist Omission of negative or weak reflections from the calculation of Rmerge This is often undocumented behavior of crystallographic data scaling merging software Examples include a elimination of a reflection from both the data set and the calculation of R merge if I lt O b conversion of I lt 0 to I 0 before the calculation of Rmerge and inclusion of this reflection in the data set the statistics of such a type are included in the Scalepack output for reasons of comparison This is the first Rmerge table in the log file not the final one c omitting reflections with I lt O from the calculation of Rmerge but inclusion of these reflections in the output data set Default sigma cutoffs set to unreasonable values like 1 This is in fact the default of the R Axis software Use of unreasonable default recommended rejection criteria in other programs These eliminate individual I s which should contribute to Rmere and yet are still statistically sound measurements Use of the eigenvalue filter to determine the overall B factor of a data set collected on a non frozen decaying crystal In this case the eigenvalue filter will calculate an overall B factor which is appropriate for the middle of the data set yet apply this to all data As a result the high resolution data will be down weighted compared to data processed with the first least decayed frame as the reference Th
252. the positive ones before averaging the result will be highly biased In Scalepack SIGMA CUTOFF is applied before averaging If one rejects all negative intensities before averaging a number of things would happen The averaged intensity would always be positive For totally random data with redundancy 8 in a shell where there was no signal there would be on average 4 positive measurements with average intensity one o This is because the 137 The HKL Manual 3 4 1 2 negative measurements had been thrown out So the average of the four remaining measurements would be about 2 o This would look like a resolution shell with a meaningful signal R merge would be always less than the R merge with negative measurements included A SIGMA CUTOFF of 1 would improve R merge even more by excluding even more valid measurements Why should this worry you Exclusion of valid measurements will deteriorate the final data set One may notice an inverse relationship between Rmerge and data quality as a function of SI GMA CUTOFF So much for using R merge as any criterion of success Even the best averaged estimate of intensity may be negative How to use negative I estimates in subsequent phasing and refinement steps The author of Scalepack suggests the following You should never convert I into F You should square Fac and compare it to I Most but not all of the crystallography programs do not do this That is life In the absenc
253. tion Indexing Indexing Lune separation In order to assign an index properly to each spot on a diffraction image the orientation of the crystal in space must be known Historically this chore was performed by adjusting the arcs of the goniometer on which the crystal was mounted until test or setting photographs showed that the crystal was aligned Usually this meant that one of the unit cell axes was perfectly aligned with the spindle axis It was then a relatively straightforward matter to index the spots on an oscillation photograph knowing the unit cell parameters the oscillation range and crystal alignment Modern data collection strategies however typically involve the collection of oscillation photographs from crystals where the orientation of the reciprocal lattice relative to the X ray beam and the film is not known ahead of time The chore of indexing has thus become one of deducing the orientation from the collected oscillation photographs rather than adjusting the position of the crystal before data collection begins There are five parameters that specify the orientation of the crystal relative to the X ray beam VERTICAL AXIS SPINDLE AXIS and CRYSTAL rotz rotx and roty values These are what we wish to deduce in our Denzo session Note that our primary goal is to deduce the orientation of the crystal relative to the X ray beam even though the orientation of the detector relative to the beam is of equal importan
254. tion is going well there is pressure to use the expensive detector and X ray beam resource efficiently The three basic questions are 1 whether to collect 2 what to collect 3 how to collect 1 Is the data worth collecting Quick scaling of a partial data set collected in the first minutes may eliminate the need to collect a full set of non derivative data Observing many diffraction spots in the image encourages one to collect a full data set however a high number of spots may be due to high mosaicity making such a data set unprocessable One image is enough to index it estimate mosaicity and note the severity of the problem caused by overlapping between the reflections If the Bragg peaks are not resolved there is no point in collecting such data despite the number of spots in the image THE HKL MANUAL 2 What range of data to collect Typically one wants to collect up to the resolution limit defined by the ratio of average intensity to o noise being about 2 The safest way to establish this is by processing a test image rather than by guessing One has to note that some space groups have inherent ambiguities in indexing which only scaling of the initial image to the previously collected data can resolve Otherwise one risks recollecting already measured indexes rather than filling in the missing data 3 How to collect data Detector The detector should be placed as far away from the sample as possible to cover
255. tion is useful for diagnosing hardware problems Detectors with problems will have dips in what would otherwise be a smoothly varying curve To exit from the A D test click on the Dte again Changes the right side scale bar so you can get an idea of where your image is on the full dynamic range of the detector Changes the lower level for the display range Basically causes the image pixels with intensities below the floor to be displayed at the lowest color darkness level so in effect it narrows the displayed dynamic range but increases the contrast The left middle right button produces a small medium large change Used in conjunction with the Frane Starts a movie by sequentially incrementing or decrementing the last digit of the file name being displayed If your file is called xtal001 osc and frame up is selected hitting will tell the display to sequentially display xtal002 osc then xtal003 o0sc etc until you stop it or run out of frames This is a neat way of screening through your frames and spotting any that have obvious problems Of course it works better on a fast computer You can do the same thing below manually with Is the opposite of Floor Up_ Increases the dynamic range being displayed by lowering the floor and again the change depends on which mouse button is pressed Colors saturated image pixels red Inverts the dark light display Changes white to black and black to white and all others in
256. tional refinement and as reference profiles in profile fitting E The intensities of all reflections determined to be weak are still measured and included in the x file They re just not used in the profile fitting You can use Prof fit R in the XdisplayF window to see which reflections are above the weak level format WEAK LEVEL value dimensionless default example WEAK LEVEL 5 00 WRITE PREDICTIONS WRITE NO PREDICTIONS This tells Denzo to write a file called hklpredictions This file contains the positions of the predicted reflections and is read by XdisplayF to update the predictions in the display Once the flag is turned on the file is updated every time a new cycle of refinement is executed This way you can follow the course of the refinement in the display program Turned off with WRITE NO PREDICTIONS format WRITE PREDICTIONS default WRITE NO PREDI CTI ONS example WRITE PREDICTIONS X BEAM Defines the distance from the edge of the data to the beam spot in mm You can use coordinates of the cursor in the green box of the XdisplayF window to find the beam position Format X BEAM value_ mm default FILM width 2 example X BEAM 98 961 Y BEAM Defines the distance from the edge of the data to the beam spot in mm Again use coordinates of the cursor in the green box of the XdisplayF window to find the beam position format Y BEAM value_ mm default FILM length 2 example Y BEAM 100 146 73 THE HKL MANUAL
257. tions may be all wrong This is 0 9A data so the polarization correction will be quite important I hope that you did not process Daresbury MAR with film rotation 90 0 or 90 0 The data would be disastrously bad Maybe you confused MAR with R axis at Daresbury R axis with horizontal spindle 3 such machines exist requires film rotation 90 0 or 90 0 Polarization number should be positive for horizontal spindle at Daresbury regardless of value of film 155 The HKL Manual rotation If you specified POLARISATION daresbury you are safe both with MAR and R axis Q When I tried to process one of virus images there was a message about sfprof too small The program is running out of memory and ignores some reflections This may happen if the profile fitting radius is unreasonably large Reduce the PROFILE FITTING RADIUS size or use the DenzoBig version of the program Overload Q If we have not overload cutoff value in the command file what does it decide as the overload value when printing the statistics after each image Default value is OK for R axis and some MAR scanners It represents the full dynamic range of possible numbers in the file Now Denzo reads overload table from MAR scanners On some old MAR scanners spiral to Cartesian conversion program incorrectly handled saturation Newer version seem to be OK Some sites may have locally modified software EMBL Hamburg MRC Cambridge check with whoever made the modification With MA
258. to be applied to display predictions from the Denzo output files To display predictions from xtal150 x file one has to type xdisp raxis xtal osc 150 xtal x xtall50 x represents the name of the file which contains a list of reflections in Denzo output format You need to click the Update Pred option in the Image Window for predictions to appear on the screen You may get a list of all supported formats by typing xdisp Formats are also listed in the Appendix Detector Specifications Using the XdisplayF 18 The frame should now be displayed on the screen For the most part the features of the display program are self explanatory The display is changed by clicking the middle mouse button or both buttons for a two button mouse once on any of the buttons above the image It is possible to zoom in and out to change the color and intensity of the image and to determine the resolution and x y coordinates in pixels and mm of any point on the image To exit from the display program click once on What follows is a more detailed description of all the options The Main Display Window Imax 55205 gt ee ag d 2 434 Fa Fa 4 hh z af an aa ah ha h t has ATER r a h a 1 9 125 8 i 3 4 ag z ai i s EY a a a a ae a aoe Su Ms yA i as HE 1 a a a zs a ga ks a a ay iy a a a an 4 ki ur a l aaa 4 a b 1 7 Processing rei F tig a a 3 i a oe i System D ie
259. toindexing routine will not use the same values for crystal rotx roty and rotz each time it is run but instead will randomly choose between equivalent values This does not matter the first time these values are chosen but if each image is indexed with different values of cRysTAL rotx roty rotz then the original index of all the reflections will be inconsistent This means that you will not be able to add partially recorded reflections correctly or even scale the data which is a major drag A third way is to incorporate a small beam search routine into the refine dat macro file above An example of such a search routine is shown below Essentially what this routine does is create a large zone within which the direct beam may be found and then exclusively search for this single reflection Note that you must actually have a spot on the frame to find if the beam stop does not have a thinned center through which the attenuated direct beam can pass then this search routine is pointless and will not find what is not there to be found Also it helps if the starting values of x BEAM and Y BEAM are reasonably good average values If this search is going to fail it is usually because the starting point for the direct beam search was too far away from the actual point It has also been known to fail if there are other small intense noise spots within the beam spot search zone Once the beam is found the box and spot sizes are reset to normal value
260. top below the maximum valid pixel value indicating a saturation of the data acquisition hardware software A correct understanding of the detector geometry is essential for accurate positional refinement Unfortunately most detectors deviate from perfect flat or cylindrical geometry These deviations are detector specific The primary sources of error include misalignment of the detector position sensors Mar R axis non planarity of the film or IP during exposure or scanning inaccuracy of the wire placement and distortions of the positional readout in multi wire proportional counters MWPC optical distortion which can also be due to a magnetic field acting upon the image intensifier in the TV or CCD based detectors E gt If the detector distortion can be parametrized then these parameters should be added to the refinement For example in the case of the spiral scanners there are two parameters describing the end position of the scanning head In the perfectly adjusted scanner these parameters would be zero In practice however they may deviate from zero by as much as 1 mm Such misalignment parameters can correlate very strongly with other detector and crystal parameters If the program does not have the ability to describe detector distortions then the other parameters such as the unit cell and crystal to detector distance will be systematically wrong With films and IPs handled manually in cassettes like at many synchrotro
261. tructure determination method 10 madnes 127 magnification 20 major axis 78 major zone 78 MAR 37 memory 17 77 146 147 154 156 MERGE Scalepack keyword 129 Minor Wladek wladek iwonka med virginia edu 4 MIPS R8000 77 misindexing 30 144 157 MONOCHROMATOR Denzo keyword 61 mosaicity 8 13 mosaicity modifier to FI T Scalepack keyword 125 very low 29 Mosaicity 153 MOSAI CI TY Denzo keyword 61 mosaicity histogram 43 MOSAICI TY Scalepack keyword 129 mouse buttons 19 24 left 20 151 158 middle 18 20 right 19 20 Multiple Native data sets 100 multiple passes of data collection 115 myexperiment dat command file 13 N negative intensities 137 Negative I s 154 NO ANOMALOUS Scalepack keyword 129 NO BEAM Denzo keyword 62 71 NO MERGE Scalepack keyword 130 NO PROFILE TEST Scalepack keyword 130 none 63 non isomorphism 89 99 non vertical spindle 42 NUMBER OF ITERATIONS Scalepack keyword 130 NUMBER OF RECORDS Denzo keyword 62 NUMBER OF ZONES Scalepack keyword 131 numerical parity 21 O Off line Image Plate 81 OMI T Denzo keyword 62 operating systems Linux 77 OSF 77 Unix 77 VMS 77 147 148 OPERATING SYSTEMS VMS 22 140 147 154 operative philosophy 77 ORIENTATION AXIS 1 Denzo keyword 70 ORIENTATION AXIS 1 Scalepack keyword 131 ORIENTATION AXIS 2 Denzo keyword 72 ORIENTATION AXIS 2 Scalepack keyword 131 Original index 130 ORI GI NAL WINDOW S
262. utoindexing Often protein crystal are polymorphic and unit cell is unexpected one It seems to me that if one has indexed the diffraction pattern from one image for example Frame 1 Then after inputing all the parameters determined from this image such as CRYSTAL rotx roty rotz etc one should be able to predict other frames for example Frame 40 with rotx adjusted accordingly However the program seems always reindex the new image and somehow ignore the input parameters I know that because the crystal symmetry would allow many different but equivalent indexings however I think consistency in indexing for successive image has its advantages in case that one wants to apply some anisotropic absorption correction The confusion comes from the misunderstanding that the rotx value should be adjusted for every frame with different starting scan value You should autoindex only one image per crystal to have consistent description of crystal orientation and you should omit the peaks search file peaks file command on subsequent images After Denzo autoindex should you have a program to output CRYSTAL rotx roty rotz if one wants to switch spindle axis and vertical axis In the case of unnamed co conspirator because odd ratio of cell axes and same frames resulted in 3 or 4 equivalent autoindex results but we could not find the location of major zone on the basis of any index results I have a P21212 diffracted to 2 2A but failed to find major z
263. utoindexing is screwed up listed in order of importance the X BEAM and Y BEAM positions They can be measured from the diffraction image using the cursor and reading off the position from the little green box in XdisplayF A direct beam shot or a good educated guess say based on the intersection of Laue streaks the symmetry of the diffraction pattern etc is often necessary At synchrotrons when the data is collected on IPs in removable cassettes and the beam wanders all over the place on the image it is very helpful to have a beam stop which has been thinned sufficiently at its center so that a tiny amount of the direct beam passes through and makes a spot on each frame whose position can be measured specification of the detector format believe it it s happened the crystal to film DI STANCE The distance should be known accurately within about 1 and can be measured with a tape measure as a last resort the CASSETTE rotx and roty values Misset CASSETTE rotx and roty values may be due to 20 angles other than zero Really knowing the detector parameters only makes sense If you want to deduce some unknowns crystal lattice and cell and orientation you need some knowns too oscillation range autoindexing parameters THE LONGEST VECTOR RESOLUTION LIMITS m WEAK LEVEL BOX and SPOT peak search INDEXING Very rarely autoindexing will fail because too large or too small an oscillation range was collected In the
264. ved and predicted reflection positions These lines are easy to spot in a low zoom window so you can easily examine any problem reflections Closes the Zoom Window Green Box on the right side of the window The maximum pixel intensity in an image Resolution of the cursor position in A Note that this is only accurate once Denzo has been launched and has read in the distance and wavelength and communicated these to the display program This means that at least one GO statement has been given in Denzo Otherwise it 1s wrong Note also that it is only accurate for flat detectors which are perpendicular to the beam Intensity at the cursor position Intensities above 2 are written by some scanners in a special manner the R Axis divides strong pixels by 8 before writing them to the image file Mar scanners create an overflow table which must be consulted to find the correct value Rest assured all Denzo versions deal accurately with strong pixel values and saturation levels Red numbers giving coordinates of the cursor in mm using the Denzo coordinate frame The first number is the x coordinate the second is the y coordinate These values can be entered into Denzo command files directly when needed for example to give the X BEAM andY BEAM position 21 THE AKL MANUAL Visualisation of the diffraction space A diffraction data set forms an image of three dimensional reciprocal space 3 d image consists of a series of two
265. which can artificially lower Rinerge generally have the effect of reducing the redundancy of the data or eliminating weaker observations In crystallography the greater the redundancy of the data the worse the Rinerges because of the correlation between I and I which reduces the Rmere The greater the redundancy the lower the correlation For two measurements with the same o the correlation is 50 SO Ryerge 18 underestimated by J2 compared to the case of no correlation Known unintentional factors which lower R merge include the following PROGRAM DESCRIPTION Data collected so that lower resolution shells where the data is strong have a higher redundancy than the higher resolution shells where the data is generally weaker This can be accomplished by collecting data on detectors where 20 0 or including data from the corners of rectangular or square image plates There is nothing wrong with using this data it will just artificially lower the Rinerge Inclusion of single measurements in the calculation Of Rmerge in one widely used program which is why a table using this erroneous calculation used to be presented in the Scalepack output Although the bug in the widely used program was unintentional it nonetheless reduced the Rmerge and this may have accounted for its longevity A second more subtle bug that reduced Ryerge prompted the introduction of the keyword BACKGROUND CORRECTION Fortunately both bugs have now been fixed but
266. wo separate measurements within a data set and the statistics that result from merging the two will reflect the differences between the I reflections and the I reflections Notice that there is no need to go through a lot of jiffies which separate the I and I data and reformat them etc Step 1 scaleano com NUMBER OF ZONES 10 number of resolution shells for statistics ESTIMATED ERROR 0 04 0 04 0 04 0 04 ESTIMATED ERROR for each resolution shell 0 04 0 04 0 04 0 04 0 04 0 04 ERROR SCALE FACTOR 1 3 multiplicative factor applied to input o SPACE GROUP P212121 REFERENCE BATCH 1 ADD PARTIALS 1 to POSTREFI NE 10 FIT crystal a 180 FIT crystal b 180 FIT crystal c 180 FIT batch rotx 180 FIT batch roty 180 FIT crystal mosaicity 1 to 180 REJECTION PROBABILITY 0 0001 expected fraction of outliers in the data if your anomalous signal is strong consider using I amp I URTE NELESI ON ie ae considered equivalent when calculating scale amp B and for statistics but output keeps them separate and they are not merged ANOMALOUS FORMAT denzo_ip SECTOR 1 to 180 FILE 1 frames Denzo myxtal x OUTPUT FILE mydataano sca output file contains h k 1 I Op I Or 112 SCENARIOS Step 2 scalecompareano com NUMBER OF ZONES 10 EST MATED ERROR 0 0 0 0 these are zero because we have already scaled the data O20 040 Us 0 0 O40 04 0 Dae D gt UaU ERROR SCALE FACTOR 1 0 NUMBER OF ITERATIONS O no scali
267. xes have been determined enter them using the commands SPINDLE AXIS 1 O 1 VERTICAL AXIS 0 2 O GO Determining crystal orientation angles CRYSTAL rotz roty rotx CRYSTAL rotx v Figure 2 The convention used in Denzo is as shown in the pictures on the next two pages CRYSTAL roty v CRYSTAL rotx v CRYSTAL CRYSTAL roty v CRYSTAL Denzo rotation conventions A simulated array of predicted reflections is shown in the center box along with the operations that shift the position of the preds to those seen in the other boxes The operation used to move the preds away from the pattern given in the center display heavy lines is given above each box The spindle axis is assumed here to be horizontal 180 degrees away from the beam stop axis which is indicated by its shadow In the case of vertical spindles the subsequent effects of crystal rotations are given in parentheses and denoted with a V 79 THE HKL MANUAL The rotations CRYSTAL rotz CRYSTAL roty and CRYSTAL rotx are best thought of as rotations expressed in degrees applied to the predicted reflections preds from the initial conditions of CRYSTAL rotz 0 0 roty 0 0 rotx 0 0 Values are entered as absolute rotations from zero For example if you have CRYSTAL rotz 23 7 and then you subsequently enter CRYSTAL rotz 15 2 the applied value of CRYSTAL rotz is 15 2 not 23 7 15 2 8 5 A
268. y madness output madnes procor ascii ascii madness output xengen urf binary output from xengen Also have to supply program with info about frame width in degrees using frame width keyword Archive binary output from Hamlin software format FORMAT modifier default denzo_ip example FORMAT denzo_ip if you use one of the binary formats that the data must be scaled on the same type of computer that created the binary files due to incompatibilities in number representations between computers FRAME WIDTH Only for URF files created by Xengen Other formats do not need this specification because Scalepack can read this information off the file header The oscillation range for each frame format FRAME WIDTH value default there is no default this input is required for this format example FRAME WIDTH 0 2 HKL MATRIX Matrix for re indexing Default unit matrix This matrix is applied to the hkl s as they are read in applies to data read in after this command K 1an 2k 3 el kr A m 0k 6 ei 1 7 4h ok oye Input is in the order 1 9 HKL MATRI X works with postrefinement nb program will not accept a matrix which has a negative determinant 127 The HKL Manual format HKL MATRIX valuell valuel2 valuel3 value21 value22 value23 etc default the default is the unit matrix example HKL MATRIX O 1 O 1 0 0 o0 transforms Akl into k h l has the effect of switching a g i and b 1 is to keep the determinant positiv
269. you can safely refine all parameters together are given below Refine all parameters when resolution A lt 1 5 for triclinic lattices lt 2 0 for trigonal tetragonal lattices Otherwise don t refine Y SCALE SKEWand DI STANCE If things are going well you should see an excellent agreement between the diffraction spots and the preds You can now extend your refinement to higher resolution Here is a macro which finishes the refinement of one frame after autoindexing refineone dat RESOLUTION LIMITS 100 2 8 don t use highest res yet FIT crystal rotx roty rotz refine crystal orientations GO GO GO RESOLUTION LIMITS 100 2 2 highest resolution FIT cell x beam y beam crossfire x y xy fit other parameters FIT cassette rotx roty 33 THE AKL MANUAL FIT distance y scale skew only refine if res A S 1 5 or 2 GO GO GO GO GO A summary of the current values for all the parameters you have refined can be obtained by typing the command LI ST Now it s time to examine the refined predictions and adjust the mosaicity and spot shape More about refinement how it works The integration of reflections requires knowledge of their indices and position The weak reflections can be found only by predictions based on the information obtained from strong reflections The autoindexing provides only the approximate orientation of the crystal thus the result may be imprecise if the initial values of the detector parameters are poorly known The least

HKL Manual

Contents

Download Pdf Manuals

Related Search

Related Contents