User Manual
Contents
1. the Download Files and choosing the file The minimized version of the Subunit B is also available for download In case only one tensor is used during the rigid body minimization algorithm all the four possible roto translated versions of Subunit B are available for download FANTEN CONVENTIONS FANTEN Convention in choosing the tensor In the general case the axis of the anisotropy tensor is not uniquely determined and in particular three equally acceptable solutions can be found FANTEN selects the solution that simultaneously satisfies the two following criteria a Xyy lt Weel lt Xzc b AXral lt 2 3 AXax FANTEN Euler angles convention Euler angles convention to be used in FANTEN can be chosen between the available ones in the General Parameters field for the Smart Interface and Rigid Body Docking or in the TENSOR ORIENTATION section from the Custom Interface The selected convention will be used by the program that after the tensor estimation will report the results in term of Euler angles according to the same convention In particular the Euler angles convention is defined by a 4 character string in which the first character defines if rotations are applied to static s or rotating r frame whereas the remaining three characters specify the axes about which the three consecutive rotations occur The resulting transformation matrix represents the rotation that brings the reference system2. 123456789012345678901234567890123456789012345678901234 6 LEU H 0 03 1 0 15 0 25 1 As a general rule PCSs attributes will be correctly interpreted if present within the following space columns Field Column resnum 1 3 resnam 4 9 atnam 9 12 exp 16 22 useless 23 24 tol 26 31 weight 32 37 t id 38 40 For the XPLOR NIH format see test_xplor pcs in the example files PCSs are reported according to the following syntax assign resid 999 and name OO resid 999 and name Z resid 999 and name X resid 999 and name Y resid 6 and name HN 0 4330 0 0000 In case of multiple datasets the PCS values belonging to the same set are identified by the residue number associated to the lines which do not contain the information about the experimental value Since no information about the weight associated to PCS values are included in the format by default a weight of 1 0 will be associated ii b RDC file can be uploaded in PARAMAGNETIC CYANA or in XPLOR NIH format For the PARAMAGNETIC CYANA format see test_cyana rdc in the example files the attributes related to RDCs are generally reported according to the following order 1 resnum1 the number of the first residue atnam1 the specific name of the atom of the first residue resnum2 the number of the second residue atnam2 the specific name of the atom of the second residue exp the acquired experimental value
3. t_id a index related the tensor tol the tolerance associated to the measurement weight the weight associated to the measurement freg the Larmor frequency for the measurement So ee a On ee a For all these fields a value should always be specified except for the attribute freq that if omitted is set by default to 700 0 In the case some not essential attributes are entirely missing the remaining ones should be reported according to this following example line for their correct interpretation 1 2 3 A 5 123456789012345678901234567890123456789012345678901234 35 N 35 H 2 600 3 2 00 0 040 700 As a general rule RDCs attributes will be correctly read if present within the following space columns Field Column resnum1 3 atnam 5 7 resnum2 9 12 atnam2 13 15 exp 17 25 t id 29 31 tol 3237 weight 41 47 freq 47 EOL For the XPLOR NIH format see test_xplor rdc in the example files RDCs are reported according to the following syntax resid 999 and name OO resid 999 and name Z resid 999 and name X resid 999 and name Y resid 6 and name N resid 6 and name HN assign 5 9590 0 5000 In case of multiple datasets the RDC values belonging to the same set are identified by the residue number associated to the lines which do not contain the information about the experimental value Since no information about the weight associated to RDCs are included in
4. the superimposing correlation plots of the calculated tensors together with the global Q factor Fig 5 Global Q factor in calculated as a usual Q factor on all the experimental data Tensor 1 Tensor 2 Tensor 3 Tensor 4 Tensor 5 Tensor6 i Tensor All Beccsccccssccccscccescccesccccescsosssooedl Total n PCS 486 data kd Download Files gt Total n RDC 149 data PCS fitting RDC fitting Qfactor 0 11 Qfactor 0 087 1 fitted RDC 2 fitted RDC 3 fitted RDC 4 fitted RDC 5 fitted RDC i lw k 6 fitted RDC Observed PCS ppm Observed RDC Hz Fig 5 Smart Interface It automatically generates a summary of all calculated tensors including the plots showing the agreement between experimental and calculated data for the different tensors When RDCs arising from partial alignment induced by external orienting media are provided the results of the calculation are given in terms of alignment tensors Fig 6 As mentioned in the methods section the size of the tensors is provided by their magnitude A and rhombicity R defined as 3 A A A F422 K a where A are the components of the A tensor in the frame where it is diagonal For the sake of completeness also the maximum RDC induced N H nuclear pair Dyp is reported see Tjandra N amp Bax A Science 278 1111 1114 D SisHoYnYuh NH 3 3 161 hy Action Panel Choose an action to perform PC
5. 179 90 244 SS 5244887 086938001 1574334 231264 2531815 55118 1574334 231264 5999370 602454 6279594 396048 0 2531815 55118 6279594 396048 754483 515516 Po y iy ais oean bossa O rosa E Jorasars Coo i psema foses pss Tensor in PDB layout HETATM 2242 ME ME A 999 61 494 4 176 20 347 1 00 4 04 ME HETATM 2243 AX ME A 999 62 243 3 834 20 914 1 00 4 04 ME HETATM 2244 AY ME A 999 62 156 4 558 19 702 1 00 4 04 ME HETATM 2245 AZ ME A 999 61 498 5 035 20 859 1 00 4 04 ME Transformation X T1 M X T2 25 648120 8 922831 6 379553 0 076142 4 176000 11 805573 0 249392 0 431965 0 866723 20 347000 Us E 58 459573 RR 0 296077 0 818135 0 492943 61 494000 gt Fitted PCS Subunit A gt Fitted PCS Subunit B gt Fitted RDC Subunit A gt Fitted RDC Subunit B Fig 12 Rigid Body Minimization Interface It provides as output the anisotropy parameter the values of the Euler angles the tensor matrices and the eigenvectors providing the main axes of the tensor together with the plots showing the agreement between experimental and back calculated data for both Subunit A and Subunit B and also the information about the applied transformation RIGID BODY MINIMIZATION Downloads The results of the calculation of the tensor together with the list of fitted and predicted data for both Subunit A and B are available for download in txt format simply clicking on
6. FANTEN web application USER MANUAL Introduction FANTEN see Rinaldelli M Carlon A et al J Biomol NMR 61 21 24 is a new user friendly web tool for the determination of the anisotropy tensors related to PCSs pseudo contact shifts and RDCs residual dipolar couplings available through the WeNMR portal http fanten enmr cerm unifi it 8080 FANTEN is mainly constituted by a first part for the upload of the structure and of the restraints Fig 1 see Input Files common to all interfaces and a second part where three different actions can be selected and which are briefly summarized as following e PCS RDC Fitting Custom It allows fully customized calculation of the tensor and for any dataset the user can define if known metal position tensor orientation and anisotropy parameters or any combination thereof see CUSTOM INTERFACE e PCS RDC Fitting Smart It allows performing tensor calculation for multiple datasets of PCSs and RDCs and for multiple models present in the PDB file Moreover it contains additional functionalities such as the generation and visualization of the isoPCS surface in case PCSs are provided and the estimation of Monte Carlo error on the tensor parameters see SMART INTERFACE e Rigid Body Minimization It allows the upload of a second structure and set of restraints and computes the roto translation to apply to the second structure to obtain the best agreeme
7. S RDC Fitting Smart S General Parameters Dataset Weight Temperature Order parameter Euler axes fix bond PCS weight RDC weight n x bon 298 0 k 0 9 szyz gt LJ lengths Tensor 1 RDC inrdc file metal n 1 In pdb file metal ME 135 A Perform Action Tensor 1 Metal n 1 in RDC file has 55 data kJ Download Files gt RDC fitting Qfactor 0 17 x 0 0 y 2 5 fo 17N17H 4 E Fitted RDC Predicted RDC Observed RDC Hz malal companen BE 0 000856 0 157045 18 203403 2 4629 2 5651 0 72419 141 11 146 97 41 493 CE 0 000005218 0 000003208 0 000008532 pO 0 000003208 0 000003063 0 000006279 D 0 000008532 0 000006279 0 000008281 CX e 885 poms fos OOO i peoa pom fosa Fig 6 Smart Interface For RDCs arising from external alignment media the size of the tensors is provided by their magnitude A the rhombicity R and the maximum Dyy value SMART INTERFACE Structures ensemble The Smart Interface permits also to perform the tensor calculation for multiple models present in the PDB file Fig 7 In this case a tensor for each model is estimated The correlation plots report the superimposing fitting obtained for the different models whose visualization can be controlled simply checking unchecking the box related to the model below the graph Moreover the tensor value the orientation and the tensor matrix will be report
8. a General Parameters Temperature Order parameter Euler axes 298 0 k 0 9 szyz O fergie Dataset Weight Subunit B Subunit A PCS weight RDC weight PCS weight RDC weight an a kiai erform gridsearch algorithm 1 0 1 0 1 0 0 193 j 9 TENSOR 1 Subunit B Subunit A Metal n Metal n a Metal in pdb ME 999 A model 1 Metal n TENSOR 2 Subunit B Subunit A Metal n Metal n 2 a Metal in pdb ME 999 A model 1 Metal n TENSOR 3 Subunit B Subunit A Metal n Metal n 3 Metal in pdb ME 999 A model 1 Metal n 3 Fig 10 Rigid Body Minimization Interface After the upload of the first molecular structure Subunit A and the related experimental datasets a second molecular structure Subunit B and the corresponding PCS RDC datasets can be provided The datasets for the two subunits corresponding to the same tensors can be paired and associated to the corresponding metals the coordinates of which must be provided in the PDB file related to the Subunit A The estimate of a roto translation of a molecular structure can be considered as a general rigid body movement problem that can be expressed according to the following least squares problem Mingeg a izl Rx d yill Eq 1 where R and d are respectively the rotation matrix and the translation vector that map the points x to the points y n 1 n In the particular case of finding the op
9. ase and represented in the correlation plot using different colours see test_cyana_predict pcs and test_cyana_predict rdc in the example files The predicted values are also reported in tabular form at the bottom of the web page together with the fitted values After the set up the calculation of the anisotropy tensor can be performed clicking on Fit Tensor Alternatively the calculation can be performed on all the tensors after their individual set up even simultaneously by clicking Fit all tensors in the Global Action field Upon checking of Fit all metals in one position the function permits also to perform the calculation of all the tensors using a unique position that will refer to the one chosen in the first tab This operation can be useful when the position of the tensor has to be optimized with the constraint that all the tensors should be placed in the same position For any selected option FANTEN performs the minimization using the Levenberg Marquardt algorithm available from SciPy library which also permits the inclusion of constraints on the parameters to be optimized in the minimization procedure A least squares approach is used to minimize the following target function TE Sd X w PESE Po y Kppc X w RDCE RDC l l where on each PCS or RDC values w is the specific weight provided in the uploaded files if present and Kpcs or kppc indicates the global weight provided t
10. bserved PCS ppm Observed RDC Hz Ax 12n 10 A 8775 293 4597 579 0 6868 2 095 0 06471 39 35 120 1 3 707 126890 56823 093 155197 499 ee 155197 499 63657 171 40038 323 Tensor in PDB layout ATOM 7900 ME 42 812 8 699 11 606 77 70 77 70 ATOM 7904 AX 43 239 9 367 16 937 77 76 77 76 ATOM 7905 AY 43 079 7 907 11 80857 77 76 77 76 ATOM 7906 AZ 43 676 8 643 12 107 77 70 77 70 gt Fitted PCS gt Fitted RDC Fig 3 Custom Interface Overview of the results of the fit performed using PCS and RDC datasets CUSTOM INTERFACE Downloads The results of the calculation of the tensor together with the list of fitted and predicted data are available for download in txt format simply clicking on the Download Files and choosing the file SMART INTERFACE If the tensor position is well known the Smart interface can be used for easy and fast calculation of several tensors Upon selection of PCS RDC Fitting Smart in the Action Panel a basic interface permits the set up of the calculation operating on the General Parameters field and on the relative weight associated to the PCSs and RDCs datasets if both present during the upload see CUSTOM INTERFACE Fig 4 According to the datasets present in the uploaded restraint files some frames are automatically created representing the tensors For each of them the corresponding set of restraint can be associated makin
11. chain if multiple are present The set up of the calculation proceeds through three different sections i METAL POSITION ii TENSOR ORIENTATION and iii ANISOTROPY VALUES i The user can operate on the metal position choosing among three distinct options e Unknown permits to fit the position of the tensor without any clue about its coordinates and thus it will perform a grid search starting from points in the proximity of the protein surface or from atoms constituting its backbone In particular the minimization is performed using 8 different starting points for the metal position corresponding to the vertices of a box containing the protein or randomly choosing 8 atoms belonging to the protein backbone In most of the cases 8 starting points are sufficient to estimated with high accuracy the position of the metal ion e Choose Metal in pdb file permits to choose as tensor position the coordinates associated to a metal present in the PDB file Moreover the tensor position can be fixed during the calculation otherwise optimized freely or imposing an upper bound on its distance from the selected metal e Close to Residue in pdb file permits to choose as tensor position whatever residue present in the PDB file and to optimize the position of the tensor starting from that freely or still imposing an upper bound on its distance from the same ii The user can operate on the tensor orientation ch
12. ed in terms of mean and standard deviation calculated from tensors ensemble However the results for the individual tensors as well as the table listing experimental data back calculated data and their difference in absolute value are available for download see SMART INTERFACE Downloads Tensor 1 Metal n 1 in PCS file has 342 data H Download Files gt PCS fitting Qfactor 0 233 Mra Vi 2 90 65 GLUH MH 4 fitted PCS Wi 7 fitted PCS WH 9 fitted PCS W 10 fitted PCS Observed PCS ppm modell_ model 2 model3 W model 4 modei5 model 6 W model7 model 8 M imodel 9 W model 10 Ay 12n 10 A 514 207 9 271 203 726 13 005 AX m 1 93852e 32 3 49502e 34 7 68029e 33 4 90267e 34 1 968 0 41097 1 1291 0 26656 2 5801 0 23963 112 76 23 547 64 693 15 273 147 83 13 73 ET 4310 813 671 183 7787 581 129 914 4964 736 PF 671183 7787 581 3893 826 4310 813 708 26 3500 396 o jna 4964 736 708 26 3500 396 228 487 3716 273 Model 1 Fitted PCS Yv gt Model 2 Fitted PCS v Model 3 Fitted PCS Model 4 Fitted PCS v Model 5 Fitted PCS v Model 6 Fitted PCS v Fig 7 Smart Interface When a structural ensemble is provided as input a tensor for each model contained in the PDB is estimated and AXax and Ay values the Euler angles and the tensor matrices are reported in terms of average and standard deviation
13. f points describing the axes orientation of all the tensors In this way a global rotation matrix and translation vector is estimated able to optimally superimpose the collections of tensors estimated for Subunits A and B Fig 11 A Subunit B rototranslated Subunit B Subunit B CY Solution 2 Solution 3 Q Subunit A Subunit B l A B Solution 1 Solution 4 Fig 11 Rigid Body Minimization In case of multiple tensors a single roto translated version of Subunit B is found panel A whereas in case of a single tensor four equally possible solutions are found panel B Special consideration requires the case in which Rigid Body Minimization is performed using only one tensor In this case due to the intrinsic ambiguity of this mathematic problem four equivalent solutions in terms of arrangement between Subunit A and Subunit B are possible Fig 11 B Thus four roto translated versions of Subunit B are generated and available for download Note roto translation is computed only when at least one PCS dataset is provided otherwise only the orientation of Subunit B will be optimized Note if arrangement of Subunits A and B needs only to be slightly optimized respect to the starting one Perform gridsearch algorithm option can be avoided In case the position of the metal for Subunit B is completely unknown the gridsearch approach is recommended RIGID BODY MINIMIZATION Results After the mi
14. g use of the t_id reported in the file as well as of the metal choosing among those present in the PDB file In case both PCSs and RDCs datasets are provided by default the tensor will be associated to both a set of PCSs and RDCs performing a joint calculation The independent calculation of the tensors using only PCS or RDC sets can be carried out simply unclicking JOINT FITTING and automatically the frame will split into two frames containing one dataset each Finally the calculation will start upon clicking Perform Action button Action Panel Choose an action to perform PCS RDC Fitting Smart Dataset Weight PCS weight RDC weight General Parameters Temperature Order parameter Euler axes fix bond 298 0 k 0 9 szyz MI lengths 10 o 1 M N H 1 020 MM CAHA 1 117 Z c N 1 329 Z c cA 1 526 Z CHA 2 144 TENSOR 1 B PCS in pcs file metaln 1 JOINT v In pdb file metal ME 701A FITTING RDC inrdcfile metaln 1 TENSOR 2 PCS inpcsfile metaln 2 In pdb file metal ME 702 A lt C JOINT FITTING Tensor 2B RDC in rdc file metal n 2 In pdb file metal ME 702A TENSOR 3 PCS in pcs file metaln 3 JOINT x i In pdb file metal ME 703 A gt FITTING RDC inrdcfile metaln 3 Fig 4 Smart Interface It permits to upload multiple PCSs and or RDCs datasets and determine the anisotropy tensors in a single step A sing
15. hrough the web interface in the case both PCSs and RDCs are fitted simultaneously The default values for the global weights for PCSs and RDCs are 1 and the ratio between the norm of vector with components equal to the experimental values of RDCs and of PCSs respectively CUSTOM INTERFACE Results When the calculation of the tensor is complete the results are shown in the bottom of the page Fig 3 For each tensor are reported e Correlation plots of the back calculated data against the observed ones together with the Q factor summarizing the goodness of the fit The inspection of each represented point can be done upon positioning of the cursor on the data e Tables showing the calculated tensor values in terms of axial and rhombic anisotropies and orientation expressed in radians and in degree according to the Same convention set up before the calculation The tensor matrix and its eigenvectors are also reported The triad of axes representing the orientation and the position of the tensors is also reported according to the PDB layout for easy copy and paste into the file e Tables listing all the experimental and back calculated data as well as their difference in absolute value Metal n 1 in PCS file has 204 data kd Download Files Metal n 1 in RDC file has 59 data PCS fitting RDC fitting Qfactor 0 119 Qfactor 0 14 E Fitted RDC E Predicted RDC ME Fitted PCS MW Predicted PCS 7 00 4 75 2 50 0 25 2 00 O
16. le anisotropy tensor in best agreement with both PCSs and RDCs can be determined at will through the joint fitting option Since metal position is known Smart Interface allows for a more efficient calculation of the anisotropy tensor which can be estimated directly by using the following equations see Bertini et al Progress in Magnetic Resonance Spectroscopy 40 249 273 1 Iz r y 2xy 2XZ 2yz PCS _ a Anr3 ZZ 2r2 es yy Jy Tiaa aa Aya 2 2 2 2Z4B Xap YAB 2 2rip 2 ORDC 3k a YAB XABYAB XABZAB asa 2 XAB ee oy o2 F Fy FY x 2AB i Tip K Tip a Tip Where _ Sis B YaYeh Ar 15kT 2nt For which the first derivate can be easily obtained and included in a Gauss Newton optimization approach As for the Custom Interface the contributions for each restraint are weighted by the specific weight present in the file and by the global weight inserted in the dedicated textbox present in the interface In case of joint estimation of the tensor using both PCSs and RDCs the tensor related to RDCs is scaled by a factor equal to the model free order parameter which can be directly modified in the General Parameters field present in the interface SMART INTERFACE Results After the calculation is complete the results will be reported as explained in Custom Interface In addiction to the tabs showing the data for each tensor the Tensor All tab is generated reporting
17. lso available for download and can be easily open and visualized using common visualization software such as PyMol and UCSF Chimera From Tensor All tab it is also possible to download the files summarizing the results obtained by the calculation of all the tensors Fitted PCS Fitted RDC Montecarlo Statistics Retained PCS and or RDC 70 N iterations 100 Calculate Ay 12n 10 A gt 5079 5 40 499 2821 8 32 047 AX m 1 91491e 31 1 52679e 33 1 06380e 31 1 20814e 33 0 61729 0 0080786 2 0395 0 0038985 0 0392 0 0045312 35 368 0 46287 116 86 0 22337 2 246 0 25962 p18 3019 5 828 799 27751 0 602 72 85933 0 854 11 Fs 27751 0 602 72 44300 5 828 799 41455 0 676 22 ps8 5933 0 854 11 41455 0 676 22 38719 0 1193 5 Fig 9 Smart Interface During the a bootstrap Monte Carlo approach for the estimation of the errors on the best fit parameters the user may decide the percentage of retained experimental data used during the calculation and the number of iterations RIGID BODY MINIMIZATION Selecting Rigid Body Minimization in the Action Panel the upload of a second structure Subunit B as well as of a second set of restraints is possible Fig 10 Similarly to the Custom interface model and chain considered during the calculation should be selected Once the data related to the Subunit B are selected clicking on Upload generates so
18. me frames similar to those discussed for the Smart Interface which easily permits the correct association of the restraints uploaded for the first structure Subunit A with those provided for the Subunit B together with the metal used as tensor position Metals used for the calculation should be included in the PDB file of Subunit A whereas in the PDB of Subunit B metals are not considered As for Custom and Smart Interfaces general parameters can be modified as well as the weight associated to PCSs and RDCs independently for Subunit A and Subunit B Rigid body minimization will be performed in mainly three steps 1 Anisotropy tensors are estimated for Subunit A according to procedure described for the Smart Interface 2 Aroto traslation is computed in such a way to find the best agreement between the experimental data provided for Subunit B and those back calculated using the anisotropy tensors estimated for Subunit A 3 Once the roto traslation has been applied to Subunit B new calculation of the tensors is performed using both sets of restraints Action Panel Choose an action to perform Rigid Body Minimization Subunit B upload Type CYANA Select pdb file Browse No file selected Select pes file Browse No file selected ta test_RBM_Cter_rotot eee San Type W test_RBM_Cterrdc A A T S alej lale CYANA Select rdc file e Browse No file selected Upload
19. nimization is performed the results for each tensor Fig 12 are reported as following e Correlation plots of the back calculated data against the observed ones together with the Q factor summarizing the goodness of the fit are shown for both Subunit A and Subunit B e Tables showing the calculated tensor values in terms of axial and rhombic anisotropies and orientation expressed in radians and in degree according to the Same convention set up before the calculation The tensor matrix and its eigenvectors are also reported The triad of axes representing the orientation and the position of the tensors is also reported according to the PDB layout for easy copy and past of the file e The transformation applied to Subunit B to obtain the best agreement with the experimental data in terms of translation vector and rotation matrix e Tables listing all the experimental and back calculated data as well as their difference in absolute value Tensor 1 Metal n 1 in PCS file has 26 data in Subunit A Download Files 7 Metal n 1 in RDC file has 15 data Subunit A Metal n 1 in PCS file has 63 data in Subunit B Metal n 1 in RDC file has 43 data Subunit B PCS fitting SubA Qf 0 011 SubB Qf 0 016 E SubA Fitted PCS MB SubB Fitted PCS RDC fitting SubA Qf 0 037 SubB Qf 0 039 SubA Fitted RDC W SubB Fitted RDC Observed RDC Hz 10295 159 4227 606 3 88118e 31 1 59377e 31 0 84912 1 0329 1 575 48 651 59
20. nt with the uploaded restraints see RIGID BODY MINIMIZATION Input Files The input files consist in i a PDB file and ii a a PCS and or ii b RDC files i The PDB file can contain different chains as well as different models It should necessarily contain at least one metal on which the tensor calculation will be performed In the case in which the position of the tensor is unknown a dummy atom should be included in the PDB file using unspecified coordinates as following HETATM 1204 LA LA A 105 0 000 000 0 000 1 00 0 00 LA ii a PCS file can be uploaded in PARAMAGNETIC CYANA or in XPLOR NIH formats For the PARAMAGNETIC CYANA format see test_cyana pcs in the example files the attributes related to PCSs are generally reported according to the following order 1 resnum the number of the residue resnam the name of the residue atnam the specific name of the atom exp the acquired experimental value useless always equal to 1 tol the tolerance associated to the measurement weight the weight associated to the measurement t_id a index related to the tensor oO Pia For all these fields a value should always be specified unless the file contains only one PCSs dataset In this case t_id field can be omitted In the case some not essential attributes are entirely missing the remaining ones should be reported according to the following example line for the their correct interpretation 1 2 3 4 5
21. obtained from the different models SMART INTERFACE JSmol web application Smart Interface offers the additional functionality for graphical visualization of the isoPCS surfaces computed for each tensor by using an integrated JSmol web application Clicking on Visualize Tensor a window opens in which the uploaded protein together with the isoPCS surface with threshold of 1 0 are graphically represented Fig 8 The isoPCS surface in cube format is also available for download jSmol Close Fig 8 Smart Interface The isoPCS surface with threshold equal to 1 ppm superimposed to the protein chain can be visualized through an integrated JSmol applet SMART INTERFACE Monte Carlo Statistics Smart Interface presents at the bottom page the possibility to perform the estimates of the error on the computed tensor parameters using a Monte Carlo bootstrap approach Fig 9 The user may decide the percentage of retained experimental data during the calculation as well as the number of iterations The tensor values orientations and tensor matrix are reported in terms of mean and standard deviation SMART INTERFACE Downloads The results of the calculation of the tensor together with the list of fitted and predicted data are available for download in txt format simply clicking on the Download Files and choosing the file Moreover in case at least one PCSs dataset is provided the isoPCS surface in cube format is a
22. onvention szyz ANISOTROPY VALUES A Unknown S GENERAL PARAMETERS PCS weight RDC weight 10 01 Global Actions Fit all metals in one position Fit all tensors Q choosen from Tensor 1 _Fit all tensors Tensor 1 Tensor 2 Tensor 3 Tensor 4 Tensor 5 Tensor 6 Tensor All DATASET Select a PCS dataset Select a model Select a chain dataset n 1 2 model n 1 chain A 3 Select a RDC dataset dataset n 1 S METAL POSITION Choose Metal in pdb file Choose Metal in pdb ME 701 A model 1 Fix position to this metal Fit position around this metal Fit position within a box around this metal of 5 0 A Select Metal residue res 701 chain A model 1 gt x y Zz AX 43 232 9 341 10 965 AY 43 107 7 934 11 033 AZ 43 670 8 648 12 116 Temperature Order parameter 2080 k 09 rel NH 1020 CA HA 1 117 C N c ca 1526 CHA 2 144 Fit Tensor Fig 2 Custom Interface It allows to fully customize the calculation of the tensor defining if known metal position tensor orientation and anisotropy parameters or any combination thereof Note if a weight of 0 0 is associated to single PCS or RDC values within the provided files these experimental data will not contribute to the calculation of the tensor However their back calculated values will be predicted in any c
23. oosing among the following options e Unknown let the tensor orientation to be freely fitted e Fix Euler angles the tensor orientation can be fixed imposing the Euler angles according to the selected convention e Fix axes from pdb the tensor orientation can be fixed selecting the tensor origin and axes orientations as defined by pseudoatoms included in the uploaded PDB see test_cyana_with_tensor pdb in the example files e Fix axes the tensor orientation can be fixed manually inserting the eigenvectors of the tensor expressed in its non diagonal matrix form iii The user can operate on the anisotropy values choosing among the following options e Unknown let the anisotropy value to be freely fitted e Fix Anisotropy values permits to fix the anisotropy values The GENERAL PARAMETERS block is used to apply different global weights to PCS and RDC dataset in case of joint calculation of the tensor and are set by default to 1 0 for PCSs and to norm PCSs norm RDCs for RDCs in such way the two datasets contribute equally during the tensor calculation Moreover the temperature expressed in Kelvin the Lipari Szabo order parameters as well as the bond length associated to different coupled nuclei can be fixed during the calculation Action Panel Choose an action to perform PCS RDC Fitting Custom 3 TENSOR ORIENTATION Fix unit axis from pdb Euler angles c
24. the format by default a weight of 1 0 will be associated However for the magnetic field the user should explicitly specify the value in the dedicated text box that will appear upon file upload The number and the name of the metal used as the position of the tensor can also be included at the end of the PCSs and RDCs file according to the following syntax 105 LINS LA or 105 LINS In this case the metal will be automatically associated to the uploaded dataset for the calculation of the tensor a MR GRID enabled web WELCOME TO FANTEN WEB PORTAL PROFILE gt gt eee Project Structure upload Select pdb file jg test2 pdb Browse No file selected Only pdb files are accepted Restraint upload Pseudo contact Shift PCS Type jg testl pcs CYANA PCS file Browse No file selected Residual Dipolar Coupling RDC Type jg testl rdc CYANA gt _ from external alignment RDC file Browse No file selected Fig 1 Interface for the upload of the PDB file and of the NMR restraints CUSTOM INTERFACE Custom interface is activated selecting PCS RDC Fitting Custom from the Action Panel Fig 2 For each set of PCSs and or RDCs dataset present in the uploaded files a different tab is automatically created on which the independent calculation of the tensors may be performed For each of them the specific dataset can be chosen together with the model and the
25. timal arrangement between two molecular structures by using PCS and RDC restraints R and d can be determined by making use of the following algorithm 1 Fitting of the tensor parameters for the Subunit A 2 Fitting of the tensor parameters and of its position for the Subunit B by making use of the anisotropy values estimated for the Subunit A 3 Two sets A and B of points are created comprising the coordinates of the tensor axes the center and the three vertices of the unit axes of the Subunits A and B such that A x4 hg Ayl B ly won nd 4 Since Eq 1 is linear with respect to the vector d the centroid of each set of points x and y can be computed and subtracted to sets A and B as following A x X X X B y 7 Yn VI thus reducing the least squares problem expressed in Eq 1 to mingea LeallRA Bile Eq 2 where RA B indicates the Frobenius norm of the matrix The problem in Eq 2 defines an Orthogonal Procrustes problem that can be solved by making use a Singular Value Decomposition SVD approach see Sch nemann PH Psychometrika 33 19 33 5 Once the center of mass has been subtracted the rotation matrix R can be easily determined as R U diag 1 1 det U V Vf where USV BA 6 Finally the translation vector d can be retrieved as d y Rx In case multiple tensors are available for the determination of the roto translation sets A and B are collections o
26. to coincide with the principal axis system of the anisotropy tensor passive rotation As example in the szyz convention where three consecutive rotations of a P and y about the fixed initial z y and z axes respectively of the reference frame occur the rotation matrix is cos q cos p siny sina cosy sinacosfsiny cosacosy sinf siny cos q cos p cosy sina siny sin q cos p cosy cosa siny sinf cosy R cos q sin fp sin q sin p cos f CONTACT For any help please contact me at carlon cerm unifi it
Download Pdf Manuals
Related Search