Version 1.1 User Manual - Laboratory of Biomolecular Modeling
Contents
1. 1 if that was a receptor atom it is unlikely to affect the final result greatly and may be neglected 2 however if MHP type assignment failed for quite a few receptor atoms 10 or for a ligand atom it 1s desirable to correct the problem and restart the calculations depending on the number of atoms for which MHP type assignment failed PLATINUM can choose to proceed or terminate in the case 2 the user will have to inspect and correct the mistakes manually before proceeding usually this happens when automatic adding of hydrogen atoms left some non standard atoms with unfilled valencies remember that PLATINUM uses full atom MHP parameterization by Viswanadhan et al 1989 and Ghose et al 1998 and hydrogens are required 26 3 Getting Help Feedback and References This web service combines all our best results obtained in the efforts to improve the performance of standard molecular docking packages It 1s absolutely free to use for Academia commercial users must ask for a licence We hope that PLATINUM will be useful in molecular modeling of interactions of ligands with receptors Please use the following reference to cite the software Timothy V Pyrkov Anton O Chugunov Nikolay A Krylov Dmitry E Nolde Roman G Efremov PLATINUM a web tool for analysis of hydrophobic hydrophilic organization of biomolecular complexes Bioinformatics 25 9 pp 1201 1202 2009 All questions about PLATINUM can be asked at the net forum2. P atom with formal positive charge metal ligation bond M metal ion A ligated atom same as acceptor in h bond stacking 1 d lt 3A d displacement h height and a including guanidinium grou 2 h lt 5A and cosa gt 0 1 for parallel ge Ae Ate eee angle between two plane 2 h lt 6A and sina gt 0 1 for T shaped fragments represented by their Siete stag centers see picture in Theory section hydrophobic At least 2 recptor hydrophobic atoms are within 7A to hydrophobic ligand atoms 2 Usage Briefly the workflow of PLATINUM 1s the following Step 1 Upload ligands and receptor in separate files Section 2 1 Step 2 Set MHP parameters Optionally mark here one of the ligands as the reference to compare all the others to this one by RMSD e g ligand pose taken from the X ray structure Then launch PLATINUM calculation Section 2 2 Step 3 Two panels appear on the results page On the Receptor panel user can set parameters for generation of 2D hydrophobicity maps for certain types of molecules lipid bilayers and a helical peptides Section 2 3 On the Ligand s panel user can sort docking solutions by one of the interaction terms Download data in a tab separated text file Section 2 4 Step 4 If desired it is possible to view distribution of hydrophobic properties for selected best solutions either by downloading files with precalculated MHP data click the floppy disk icon la or in Jmol a
3. 25 WARNING unidentified atom type found trying to detect atom types automatically if atom type is given other than a sensible element e g Du dummy atoms Lp lone pairs Any any atoms Het heteroatoms etc which are found in abundance in mol2 format then PLATINUM will try to detect atom type automatically as it is done for gro or pdb note that if unsuccessful PLATINUM will omit this atom lone pairs are omitted anyway WARNING ignored atom if the program encounters an atom for which element type cannot be identified or a pseudo atom Lp Du etc it 1s omitted ERROR file contains no atoms file likely to be not of the corresponding molecular format program will terminate inspect the corresponding file manually ERROR cannot open input file if you have received such message then there are some problems with our web server please tell us so that we could fix the problem WARNING ligand file contains metal atom will be omitted only receptor structure can contain a metal ion if metal ion is necessary for analysis of molecular properties than it should be submitted as part of the receptor molecular file WARNING platinum MHP TYPE ASSIGNMENT FAILED IGNORED ERROR platinum MHP TYPE ASSIGNMENT FAILED PLATINUM automatically identifies the valencies of atoms and bond orders and based on these data assignes atomic MHP constants if it fails to do so for any atom its MHP constant will be set to zero two ways are possible
4. 13 Distance function exponential exp r 2 exponential exp r hyperbolic Fermi like Since MHP is an empirical approach no exact distance dependent decay function is known like it is for electrostatic or van der Waals energy Several different approaches have been described in literature The most popular is the exponential function N MHP gt feo i 1 where MHP is the hydrophobicity potential at point j N is the number of atoms r is the distance between point j and atom i in A and a has the dimension of A Usually a 1 Fauchere et al 1988 or a 2 A Gaillard et al 1994 default in PLATINUM are used Alternatively a hyperbolic function was proposed by analogy with the electrostatic potential energy It is calculated according to the following formula Audry et al 1986 MHP 5 fi J 11 ar where a 1 A The unit was added to r in the denominator to avoid infinite values in the vicinity of atom Later it was shown that this approach works well for small molecules but for protein it results in overaveraging of MHP on the surface due to large contribution of inner atoms and therefore a more rapidly decaying dependence is needed Finally one can use averaging of atomic hydrophobicity constants in the vicinity of point j with a cut off function smoothed as a Fermi like potential Heiden et al 1993 unp i a tee J N 1 ta e ar 4 1 where a 1 A Here only atoms tha
5. of MHP calculation on grid are The output for InsightII is saved to files ligfilename_mhplig gdf formatted grid of ligand MHP ligfilename_mhprec gdf formatted grid of receptor MHP if the option was requested ligfilename_mhplig sh csh script to call unformat phi utility on Silicon Graphics platinum spect InsightII color spectrum for MHP The output for MoIMol is saved to files ligfilename_mhplig pot formatted grid of ligand MHP ligfilename_mhprec pot formatted grid of receptor MHP if the option was requested ligfilename_mhplig mac MolMol macros 24 2 8 Processing WARNING and ERROR messages When calculations finished there can appear WARNING or ERROR messages on the Results page WARNING means minor mistakes that can be treated by the program and proceed calculating ERROR means an unavoidable mistake making further calculations senseless and require user intervention Below is given the list of WARNINGs and ERRORs WARNING CONECT records will be ignored CONECT records in pdb files are not processed by PLATINUM bonds in pdb and gro files are defined solely based on distance between atoms if one wishes to work with complex or nonstandard compounds it is better to use mol2 or sdf file formats where bonds are given explicitly WARNING bad file format omitting string pdb and gro files have fixed lengths for each field atom name coordinates etc if an atom string is too short it canno
6. or hydrophilic positive or negative respectively constant based on molecular topology Then hydrophobic hydrophilic properties of a molecule can be calculated on its surface and compared to the properties of neighboring molecules to estimate the hydrophobic effect MHP at point j is a sum of atomic MHP constants f decaying with the distance rj To estimate the hydrophobic hydrophilic complementarity of a ligand to the receptor binding site PLATINUM calculates hydrophobic hydrophilic properties of both the ligand and its environment on the ligand molecular surface the interface While for ligand it is rather straightforward to do some issues arise when its environment is considered Metal ions present in the active site of many enzymes lack parameterization for MHP By default PLATINUM assignes them a constant of 1 37 corresponding to experimental ogP of a water molecule Solvent except reacting water molecules is omitted in the docking process leaving an empty space around the ligand PLATINUM surrounds the ligand by a rectangular grid of water hydrophilic charges Fig 1 2 Figure 1 2 Grid simulating hydrophilic properties of water which is implemented in PLATINUM Grid nodes fill the empty space around ligand dark molecule bound in the active site of receptor light molecule represented by ribbon The parametes of the grad are grid step 2 A grid node hydrophilic constant 0 38 By complementarity of h
7. or lower layer of lipids we assume that the membrane lies in XY plane of a rectangular box starting from point X Y Z 0 0 0 Residues range only for a helical peptides The quality of the 2D cylindrical map is good only for a helices close to ideal geometry If the peptide has unstructured tails it is recommended to manually select the residues that form the a helical region If the o helix has a kink it is recommended to build maps for each a helical region separately Ligand projection All ligand molecules will be projected onto the surface along the Z axis for lipid membranes and along the radius for cylindrical maps of a helices Ligands are depicted as a dotted area Shadow fill ligand molecule is projected see example in Fig 2 4 Cut off only parts of ligand closer than cut off dictance to the membrane are projected DOPS_bilayer MHP 62 48 Figure 2 4 An example of 2D hydrophobicity map of a dioleoylphosphatidylserine DOPS layer A membrane active peptide approaching the membrane uploaded as ligand is projected onto the map in shadow mode depicted with crosses 50 55 45 Y A 15 20 25 30 35 10 o 5 10 15 20 25 30 35 40 45 50 55 60 62 89 17 2 4 Scoring the Results of Docking On the Ligand s panel user can sort docking solutions by one of the interaction terms caculated by PLATINUM for each ligand pose Fig 2 5 If a reference ligand was provided there appear co
8. the front panel Fig 2 1 Here the reference ligand should also be listed in the ligand section Step 1 Upload files Figure 2 1 The upload files menu Here the Select ligand file s to view their hydrophobic properties add a si ini aa ia iv cha aS oan aie receptor and the ligand files can be uploaded If receptor is a lipid bilayer or an o helix 2D hydrophobicity maps can to t h e PL A TINUM server In t h is exam p I et h e be calculated for it File size limit 3 Mb first ligand file is extracted from the X ray Please note If Adobe Flash 9 not 10 is not installed then use simple upload mode structure IVFP complex of Ca ATPase with Receptor ATP analogue the others are docking poses ee on At the next step the file 1VFP_atp pdb will be 3 marked as the reference structure All docking rE E E poses will be compared to it by the RMSD gold_soin_atp_m1_1 mol2 4 75 KB values gold_soln_atp_m1_2 mol2 4 75 KB gold_soln_atp_m1_3 mol2 4 75 KB gold_soln_atp_m1_4 mol2 4 75 KB gold_soin_atp_m1_5 mol2 4 75 KB Upload files 11 Important issues Remember that receptor and ligand molecules should be submitted in separate files Parameterization of atomic hydrophobicity constants requires that all atoms are protonated By default PLATINUM web server uses Adobe Flash Player 9 to upload files version 10 does not work If Adobe Flash Player 9 is not installed on the com
9. 0 MHP offset 0 03 more hydrophobic Figure 2 3 Correction of tabulated atomic hydrophobicity hydrophilicity constants may give more realistic picture This is illustrated with ATP a ligand representing a wide class of nucleobase containing ligand compounds Aromatic purine ring usually tends to form hydrophobic contacts with protein residues and indeed has less hydrophilic MHP values than the ribose and phosphate moieties However a moderate shift of atomic constants to the hydrophobic range provides a more clear picture right in comparison with left providing the basis for numerical assessment of hydrophobic complementarity Also the same shift of atomic constants does not alter the overall picture for protein molecule and is therefore acceptable 15 Dot density VeryLow 1 5 dot A Low 5 dot A Medium 16 dot A High 60 dot A This parameter does not change much the overall picture However higher density makes the result more robust although it increases the elapsed time to process molecules It is recommended to use Low or Medium density for single calculations and VeryLow or Low density for analysis of more than 10 100 ligand poses Ignore MAP assignment errors PLATINUM automatically identifies the valencies of atoms and bond orders and based on these data assigns atomic MHP constants By default if it fails to do so for any atom in a ligand the error message is returned and no further calculations ar
10. MIPA UNUM User Manual Contents Oe TANTO CC TIO Ii arsenal actos oror Ea Nr OAO NOE RE E 2 ge DNC OF E TAE E A N E N PA E E eed A A EA E ET 3 FEDO Kol OAV CO gente a a E a ase O 3 1 2 Molecular Hydrophobicity Potential MHP i ccccssseeeeeeseeeeeeeeeeeeeeeeeeaaaas 4 1 3 Stacking and Aromatic Interactions ssseeseeeeeeeeecccccessncennsssssssessseeeereteeees 6 Ie4 Pyro Get BONG Seinai a A A ES 7 FRM Dant Reed leh P eee oer ce tec erate as even ferent a vee eee omer ee eee 8 eg NII AES AAAA EEE A os uc ates as ata de ae eet AEAEE 10 2 1 Wp load ie NI OIE CUNES icine onee an ee ea raa aaa e E eT a ANCES AOSE 11 2AM H P Parameter oen a O E 13 23s ZD H Y ArODHODICIY MAPS asica A vonexanacisdleadeasaebies 17 2A SCOrme the RESUS OF DOCKING sersan rr na T N Aa 18 2 5 Visualization of Hydrophobic Hydrophilic Properties in Jmol eee 20 2 6 Saving Data to View MHP in a Molecular Visualization Application 00 21 2 7 Format of Output Files for Visualizing MHP cc eeesssseesseeeeeeeeceeeeeeeeeeeeaaaaas 23 2 8 Processing WARNING and ERROR messages ccccssseseesssssseeececeeeeeeeeeeeeeaaaaas 25 5 Getting Melp F coedback and Referens sinia ee aea E EE EEE 27 GC Le Cl aN Be ALLL Ceo a E E N cone E neat 28 0 Introduction PLATINUM Protein Ligand ATtractions Investigation NUMerically web service is designed for analysis of hydrophobic hydrophobic jrganization of biomolecula
11. NS 66 288 298 2007 Pyrkov TV Priestle JP Jacoby E Efremov RG Ligand specific scoring functions Improved ranking of docking solutions SAR QSAR Environ Res 19 91 99 2008 Schrodinger LLC Maestro User Manual 2007 Testa B Carrupt PA Gaillard P Billois F Weber P Lipophilicity in molecular modeling Pharm Res 13 335 343 1996 Viswanadhan VN Ghose AK Revankar GR Robins RK Atomic physicochemical parameters for three dimensional structure directed quantitative structure activity relationships 4 Additional parameters for hydrophobic and dispersive interactions and their application for an automated superposition of certain naturally occuring nucleoside antibiotics J Chem Inf Comput Sci 29 163 172 1989 28
12. PLATINUM also rotates the hydrogen atom seeking for where a Qapn Qidea best geometry ideal 0 Figure 1 5 Scheme of geometrical parameters used to describe a hydrogen bond 1 5 RMSD and TcIFP Root mean square deviation RMSD and Tanimoto coefficient of interaction fingerprints TcIFP are the measures of similarity between two distinct docking poses Using these criteria all docking solutions can be clustered into groups to reduce redundancy or compared to the correct crystallographic structure when testing the performance of a docking algorithm RMSD is the most common way to compare docking poses RMSD between poses a and b is calculated over 3D coordinates of all N heavy atoms of ligand by the following equation Na xt xP yf yb 22 2 N RMSD For two completely similar positions of ligand in 3D space RMSD 0 Usually in docking applications RMSD S 2 A is considered acceptable i e if docking yields ligand poses within this cutoff of RMSD from crystallographic or NMR structure for most of the reference receptor ligand complexes it is likely to produce reliable predictions for others IFP represents a more complex notion and has not yet achieved wide use The basic idea 1s that RMSD does not always provide complete information about the similarity of orientation of a ligand in the binding site Thus in some cases even low RMSD may correspond to distinctly different intermolecular co
13. arge 0 0100 charge Molecular file This is perhaps the most universal format of output Both mol2 and pdb file formats are understood by most molecular modeling software and almost always can be used to color the molecular surface by the value of electric charge and B factor for pdb The parameters of MHP calculation on surface are the same as for grids and surfaces Additional option 1s the selection to write down tabulated atomic MHP constants or projection from molecular surface calculated inwardly by PLATINUM The second variant provides smoothed and more realistic view of molecular MHP and implies calculation of MHP on ligand surface in the same way as if you have chosen to save molecular surface in the previous point and afterwards averaging and projection back to ligand atoms If you use e g PYMOL then type the following command to color the molecule by MHP a cmd spectrum b cyan white yellow selection all or use green white yellow or red white blue color schemes instead b for PYMOL versions 1 and higher a color_b py script can be used color b mode hist gradient user user_rgb 0 0 5 1 1 1 1 0 5 0 25 0 color b mode hist gradient user user_rgb 0 0 1 1 1 1 1 1 0 color b mode hist gradient user user_rgb 1 0 0 1 1 1 0 0 1 The output for tabulated atomic MHP constants is saved to files ligfilename_mhplig pdb or mol2 file of ligand molecule ligfilename_mhprec pdb
14. culations of hydrophobic interactions without PLATINUM To use this option simply click on the floppy disk icon kal just next to the ligand pose you wish to view A menu appears where you can choose the format of output data and readjust some parameters of MHP calculation Fig 2 7 You can save data in a number of different formats depending on which molecular visualizing application you prefer to use and what possibilities 1t provides This section gives a brief overview of available formats Detailed discription of output files is given in the Section 2 7 Format of Output Files for Visualizing MHP gald_soln_atp_rmt_2 EJ Save as Molecular surface Dot density Medium MHP table New 1996 Distance function exponential expira Color gradient Brown Cyan Calculate MHF distribution fram the protein as well iwo files Save sudace Figure 2 7 Menu for visualization of molecular hydrophobic hydrophilic properties Most of the MHP parameters are the same as for processing the results of docking Section 2 2 Fig 2 2 A new option here is the Color scheme Different color schemes may be used Fig 2 8 Brown Cyan is the default This parameter defines the colors corresponding to hydrophobic and hydrophilic parts of molecules and may affect some output data Besides ligand MHP one can request calculation of MHP of its environment receptor and water To do that mark the check box Calculat
15. e MHP distribution from the protein as well Then additional file will be provided 21 MHP constants Save a table of atomic MHP constants PLATINUM provides the possibility to facilitate assignment of these constants which otherwise would consume much effort and time to do it manually Can be used for further private calculations Molecular file Save results either to mol2 or pdb format These files can be later loaded into a molecular visualizing program e g PYMOL and colored according to MHP data written to their occupancy and B factor columns both columns will contain the same numbers In addition there is an option to save either tabulated atomic hydrophobicity constants or surface MHP projected back onto atomic centers The latter provides smoothed and more realistic view of molecular MHP Molecular surface Save data either in the format of InsightII Molecular Simulations Inc 2000 surface or as a pdb file as a set of atoms Grid Save data in the format of either InsightII or MolMol Koradi et al 1996 In both cases additional files spectrum and scripts will be provided to facilitate the procedure of visualizing molecular properties Figure 2 8 Color schemes for MHP representation Three different variants are proposed Brown Cyan brown oil colored hydrophobic and cyan water colored hydrophilic properties Yellow Green another common scheme Blue Red blue carbon nitrogen at
16. e performed PLATINUM is more forgiving to errors in the receptor molecule in assumption that if only a minor portion of atoms have mistakes gt 10 they are likely to reside far from the active site and will not influence the final result However if you are sure that the errors in one or even more atoms of the ligand molecule may be neglected then mark the Ignore errors check box This will force the MHP calculations anyway This can be done e g when a peptidic ligand of gt 10 is not properly capped and manual correction of such mistake may become a needless routine After all the parameters are set and reference ligand selected press the button Launch Platinum Then your submission is queued and as soon as PLATINUM finishes the results page is displayed the page will refresh every 30 seconds until PLATINUM finishes If e mail was provided a notification is sent to this address 16 2 3 2D Hydrophobicity Maps On the Receptor panel user can set parameters for generation of 2D hydrophobicity maps for certain types of molecules lipid bilayers and a helical peptides 2D maps for lipid bilayers can only be built for gro file format because the size of the box is required If the uploaded structure is not a lipid bilayer or an a helix maps will be built anyway but it will make no sense Hydrophobicity map options Color scheme Same as in Section 3 5 Brown Cyan Yellow Green Blue Red Layer only for lipid bilayers Upper
17. eraction terms a H bonds likely to be a fractional number b Lipophilic match surface Sz A see Fig 2 3 c Hydrophilic match surface Syy A d Sburied Sux Sta SHL SHH A e Sioa StL SrH Sux Suu Sre Sur f Fraction of lipophilic and hydrophilic match Sz SHH Stotai g Fraction of lipophilic match 2 x 77 2 Sz StH Spy San h Stacking with aromatic rings see Theory Section 2 3 1 Stacking with guanidinium groups Clicking on the name of each interaction term will sort the docking poses by its value Click on Ligand name to restore the initial view You can sort the list according to one of the listed terms If any term is equal to zero for each ligand pose it will not appear in the table 18 Generally Matchl and Match2 terms are recommended for scoring and are of similar efficiency the choice between them might depend on the user s preference and intuition Other terms are given as the additional information which may be useful in some cases Scoring by hydrophobic complementarity Sz 1s recommended as default however the user is free to inspect other variants This term has been optimized to use with goldscore scoring function for rescoring results of docking Pyrkov et al 2008 obtained with the program GOLD Jones et al 1995 This number should be simply multiplied by the factor of 0 015 and added to the corresponding goldscore val
18. ig 2 2 Below is the list of these parameters with discussion of the effect they may exert on the final result Options PLATINUM options MHP table Distance function New Ghose 1998 i exponential expi r 2i MHP offset 1 5 44 57 0 00 Dot density Your email IF you enter your email a letter with your job ID will be sent otherwise you should stay at this site and wait until the program finishes pyrkovG nirnir ru Figure 2 2 The MHP parameters tab You can also provide your email address to receive URL for the results page on the PLATINUM server by mail MHP table New Ghose 1998 Obsolete Viswanadhan 1989 The parameterization of atomic hydrophobic hydrophilic constants used in PLATINUM implies a full atom model make sure that hydrogen atoms were added prior uploading files Two variants of this parameterization exist The former one Viswanadhan et al 1989 and the upgraded Ghose et al 1998 The resulting picture may differ for these two parameterizations The major changes in the new table relate more realistic negative hydrophilic constants for some heteroatom types particularly oxygen Therefore it is recommended to use the new version although the previous one was also kept in the program mainly for comparison purposes For protein residues there will be little difference however for low molecular weight compounds the difference between these two parameterizations may be prominent
19. ion 3 3 Scoring the results of docking 1 3 Stacking and Aromatic Interactions Aromatic fragments often form stacking and T shaped contacts which to our view are not always paid due attention in many scoring functions The most popular example is DNA where parallel stacking between nitrogen bases contributes much to the stability of a macromolecule While this type of interactions is not as widespread as e g h bonds or hydrophobic contacts the thing may be quite different in case of some specific types of ligands Thus we have shown that stacking plays important role in recognition of ATP Adenosine triphosphate by proteins Pyrkov et al 2007 Figure 1 4 Scheme of geometrical parameters used to describe a stacking contact between two aromatic rings Displacement d and height h are calculated for the center of one aromctic ring relative to another ring s plane Angle a is calculated as the angle between the normal vectors of both rings Currently PLATINUM implements a simple method to detect a stacking contact between a ligand and a receptor based on geometrical parameters Stacking in PLATINUM is estimated by the following formulas S Sa d S h X Sala where d and h are respectively displacement and height of the center of one aromatic ring relative to each other and a is the angle between their planes Fig 1 4 Sa max cos a sin a where the particular form of S a defines whether we
20. lumns with RMSD and IFP see Theory Section 1 5 If There are warnings caption appears user can click it to see the list of problems with atom type assignment The Save this table in tab delimited format button is below the table Clicking here saves the table in a tab separated text file on a local disk for further analysis User can change MHP parameters and recalculate hydrophobic properties To do that press the Change settings amp restart caption in the upper right corner Or press the Start a new task caption to start PLATINUM from the very beginning Ligand s Reference feand WFF atp pdb Ligand name rmsd IFp H bonds SuL SHH Shuriea Swi Match Match Stack Stack Gua_o gold_soln_atp_m1_ 1 94 0 714 5 55 5 25 248 80 289 11 322 97 0 7866 208 Oo 804 0 gold_soln_ate_rnd_4 m 1 21 0 684 4 16 0 06 242 53 279 18 309 37 0 7842 Oo 0034 1 742 QO gold _solr_ate_rnd_4 m 1 19 0 632 4 27 0 25 244 18 280 63 313 54 O 7ro6 0 0128 1 891 QO WFFP alp m 0 00 1 5 46 0 95 2448 23 290 06 319 68 0 7794 0 O424 1 935 S gold_soln_atp_m1_3 al 3 89 0 55 1 87 1 77 229 30 252 59 311 52 0 7418 0 1083 1 659 QO gold_soln_atp_m1_5 m 8 16 O 2408 2 04 0 00 234 87 287 97 335 25 7006 121 H Save this table in tab delimited format Figure 2 5 Interaction terms calculated by PLATINUM for ligand docking poses Reference ligand is marked with orange color Here is the explanation of the int
21. ntact pattern and vice versa One way to treat this problem may be using some quantitative measure of similarity of contacts with receptor such as hydrogen bonds salt bridges etc And IFP provides such measure Asp8 amp 6 Gin131 Leul34 Heo CHEE a h bond main chain stacking NS ao0nona Pa h bond hydrophobic salt metal bridge Ligation Figure 1 6 An example of IFP bit string Inhibitor staurosporin is shown in Cyclin Dependent Kinase 2 active site PDB code 14Q1 Interactions of selected residues are shown as they are encoded into an IFP bit string Introduced recently Deng et al 2004 IFP represents a conversion of 3D pattern of contacts between ligand and receptor into 1D bit string Here a set of bits corresponding to different intermolecular contacts is assigned to each receptor residue in the vicinity of the ligand binding site Fig 1 6 in PLATINUM these bits are different from those in the original implementation by Deng et al 2004 see below When a contact with a particular residue 1s observed e g a hydrogen bond is formed the corresponding bit is raised And two bit strings coding two docking poses can be compared now indicating whether they are similar or not To compare the strings A and B coding two ligand poses a Tanimoto coefficient 1s used IAN BI TcIFP AUB Here ANB is the number of raised bits common to both poses and AUB is the number of bits raised for either ligand A or B Consequntl
22. observe parallel cosa gt 0 5 or T shaped edge to face sin a gt 0 5 stacking 1 0 h lt h where h 4 0 A and h 5 0 A for parallel stacking S h 4 ho h ho hy h lt h lt ho and h 5 0 A and h 6 0 A for edge to face 0 0 ho lt h stacking 1 0 d lt d where d 2 0 A and d 3 0 A for any type of Sa d 4 do d do d1 d lt d lt d stacking arrangement The parameters of stacking 0 0 do lt d were derived from the statistical analysis of contacts of nucleobases with aromatic amino acids in proteins Pyrkov et al 2007 For each ring of a conjugated aromatic cycle stacking is calculated separately and then summed for e g naphtalene the value can be greater than 1 0 Flat guanidinium groups which can be found in Arg side chain are treated in the same way as aromatic rings 1 4 Hydrogen Bonds Hydrogen bond are also identified simply by geometrical criteria according to the following formulas B B rap X Ba apy where rap is the distance between the hydrogen donor and acceptor and dapx is the angle betwee acceptor donor and the hydrogen atom Fig 1 5 1 0 Tap STI where r 3 2 A and ry 3 4 A B rap fon Tap 1 T lt Tap lt To 0 0 S YAp 1 0 asa where a 0 and a 20 when donor is B Qapy ie amp amp o amp 1 A lt a lt a nitrogen and a 20 and a 40 when 0 0 A lt a donor is oxygen in this case
23. ods An analysis of ALOGP and CLOGP methods J Phys Chem A 102 3762 3772 1998 Heiden W Moeckel G Brickmann J A new approach to the display of local lipophilicity hydrophilicity mapped on molecular surfaces J Comput Aided Mol Des 7 503 514 1993 Jones G Willett P Glen RC Molecular Recognition of Receptor Sites Using a Genetic Algorithm with a Description of Desolvation J Mol Biol 245 43 53 1995 Kitchen DB Decornez H Furr JR Bajorath J Docking and scoring in virtual screening for drug discovery Methods and applications Nat Rev Drug Discov 3 935 949 2004 Koradi R Billeter M Wuthrich K MOLMOL a program for display and analysis of macromolecular structures J Mol Graph 14 51 55 1996 Meyer EA Castellano RK Diederich F Interactions with aromatic rings in chemical and biological recognition Angew Chem Int Ed Engl 42 1210 1250 2003 Moitessier N Englebienne P Lee D Lawandi J Corbeil CR Towards the development of universal fast and highly accurate docking scoring methods A long way to go Br J Pharmacol 153 S7 S26 2008 Molecular Simulations Inc InsightII User Guide MSI San Diego CA 2000 Pyrkov TV Efremov RG A fragment based scoring function to re rank ATP docking results Int J Mol Sci 8 1083 1094 2007a Pyrkov TV Kosinsky YA Arseniev AS Priestle JP Jacoby E Efremov RG Complementarity of hydrophobic properties in ATP protein binding A new criterion to rank docking solutions PROTEI
24. olecular recognition and sustaining 3D structure of complex molecular systems in a living cell e g lipid bilayers lipid protein complexes The force driving apolar substancies to avoid contacts with polar solvent water results in effective hydrophobic lipophilic interaction and has very complex physical nature To describe this effect in molecular biology empirical methods are used which are mostly based on experimental data on partitioning of organic compounds between polar and apolar environments see review for details Efremov et al 2007 A popular approach is to use the formalism of Molecular Hydrophobic Potential MHP a k a Molecular Lipophilic Potential MLP Each atom is assigned a hydrophobic positive or hydrophilic negative constant derived from experimental data PLATINUM uses parameterization on octanol water partition data from two sources Ghose et al 1998 and Viswanadhan et al 1989 Then the hydrophobic properties can be calculated on molecular surfaces or on a 3D grid Fig 1 1 Exponential decay is most popular Fauchere et al 1988 Gaillard et al 1994 however hyperbolic Audry et al 1986 or Fermi like cut off Heiden et al 1993 functions are also applied in some studies and are implemented in PLATINUM Figure 1 1 The concept of Molecular Hydrophobicity Potential MHP Let us consider for instance Russian vodka Each atom in a molecule ethanol or water is assigned a tabulated empirical hydrophobic
25. om hydrophobic and red water oxygen atom hydrophilic properties Choise of the color scheme may influence the format of output for InsightI surface and grid and MolMol grid 22 2 7 Format of Output Files for Visualizing MHP Below each option will be described with its parameters in detail Results for each option are written to files that should be saved on local disk Filename conventions are the following Let us say that visualization was requested for ligand named ligfilename mol2 or ligfilename sdf etc Then any file with the results of PLATINUM calculation will start with this name ligfilename_ with the appropriate suffix and extension MHP constants Assignment of hydrophobic constants to atoms of a molecule is based on sophisticated molecular topology For e g protein residues these can be easily tabulated but when ligands are processed manual constants assignment becomes a problem that may be much effort and time consuming PLATINUM makes such assignment in an automatic way These data may be used to perform your own calculations of MHP in case there is no appropriate option in PLATINUM The output is a text file ligfilename_mhptab txt The format of a line in this file is Atom Atom Atom Residue Residue Atom Atom CVFF number name Sybyl number name MHP MHP electrostatic 7 a constant O 50 ATOM 26 ae 1 ARG Eua 50 0 1036 ne 0 3278 aa a e aron 26 cs 2 TRP mhz 8 0 6805 ch
26. or mol2 file of receptor molecule if the option was requested The output for averaged surface MHP 1s saved to files ligfilename_mhplig pdb or mol2 file of ligand molecule with self MHP ligfilename_mhprec pdb or mol2 file of ligand molecule with receptor MHP if the option was requested 23 Molecular surface These surfaces may be used to visualize molecular MHP in InsightII special format surface dots or in any other molecular visualizers surface dots written as list of atoms in pdb file format The parameters of MHP calculation on surface are the same as for grids The output in pdb file format is saved to files ligfilename_mhplig surf pdb and ligfilename_mhprec surf pdb if this option was requested MHP values at surface dots are written to charge and B factor columns The output for InsightIl is saved to files lgfilename_mhplig surf and ligfilename_mhprec surf if this option was requested Also file ligfilename_mhplig pdb is provided with ligand molecule which should be loaded in InsightII to read the surfaces The format of surface file implies that two first lines contain special information Next lines are of the following format x dot coordinate y dot coordinate z dot coordinate color R G B 255 MHP value Grid These grids may be used to visualize molecular MHP on any surface in InsightII and MolIMol http hugin ethz ch wuthrich software molmol molecular visualizers The parameters
27. or via e mail to pyrkov nnrr ru Tim Pyrkov scientific concept amp programming volster nmr ru Anton Chugunov web design amp programming efremov nnr ru Roman Efremov head of the lab The authors will be grateful for any remarks on conceptual issues or suggestions on technical improvements of the service 21 4 Cited Literature Audry E Dubost JP Colleter JC Dallet P A new approach to structure activity relations The molecular lipophilicity potential Eur J Med Chem 21 71 72 1986 Berendsen HJC van der Spoel D van Drunen R GROMACS Comp Phys Comm 91 43 56 1995 DeLano WL The PyMOL Molecular Graphics System DeLano Scientific Palo Alto CA USA 2002 Deng Z Chuaqui C Singh J Structural interaction fingerprint SIFt A novel method for analyzing three dimensional protein ligand binding interactions J Med Chem 47 337 344 2004 Efremov RG Chugunov AO Pyrkov TV Priestle JP Arseniev AS Jacoby E Molecular lipophilicity in protein modeling and drug design Curr Med Chem 14 392 415 2007 Fauchere JL Quarendon P Kaettere L Estimating and representing hydrophobicity potential J Mol Graph 6 203 206 1988 Gaillard P Carrupt PA Testa B Boudon A Molecular lipophilicity potential a tool in 3D QSAR Method and applications J Comput Aided Mol Des 8 83 96 1994 Ghose AK Viswanadhan VN Wendoloski JJ Prediction of hydrophobic lipophilic properties of small organic molecules using fragmental meth
28. pplet on the PLATINUM web server click the molecule icon Sections 2 5 2 7 Important The results of PLATINUM calculations are stored for 30 days and are accessible via the unique uid code To get these results press the Goto results page caption in the upper right corner on the PLATINUM main page A dialog appears where the uid code should be entered This code is sent you by e mail if it was provided or you shoud copy and save the URL when you see the results page for the first time 10 2 1 Uploading Molecules PLATINUM supports input in five popular molecular file formats pdb mol2 sdf gro used in the GROMACS package for molecular dynamics Berendsen et al 1995 and mae used in molecular modeling package Maestro Schrodinger LLC 2007 PLATINUM does not perform ligand docking or any geometry optimization it only uses given 3D coordinates of biological molecules supplied by user Any file can be uploaded as either receptor or ligand Molecules in such files will be treated in different ways All calculations hydrophobic properties interactions with receptor etc are performed for ligands Receptor is only an environment for ligands If multiple ligands were uploaded one of them can be selected as the reference ligand to calculate RMSD and Tc IFP see Theory with other ligands However 2D hydrophobicity maps are generated for molecules or molecular systems uploaded as receptor To upload files use
29. puter one can use stadard upload file form click simple mode which however may be somewhat boring when many ligands are to be uploaded Java applet is required to be installed for on line visualization of molecular hydrophobic hydrophilic properties The size of the molecules is limited to 5 Mb which roughly corrsponds to a molecule of 40 000 atoms in mol2 file and is obviously suitable for most practical applications Any ligand file can contain more than one molecule but this is not recommended If docking poses are uploaded in a single file PLATINUM will split the file and rename all the poses adding numerical index Reference ligand can be selected at the next step You can mark any one of the entries as reference not mandatory the first one Only heavy atoms are used in calculation of RMSD and IFP RMSD will be set to 1 0 and IFP to 1 0 if either reference ligand was not marked or PLATINUM did not manage to identify corresponding atoms in ligands these are atoms with identical atom name and residue name and number 12 2 2 MHP Parameters PLATINUM was mainly designed to facilitate analysis and visualization of hydrophobic hydrophilic properties of biological molecules Some software packages provide tools to calculate these properties using the same formalism of MHP However there is often required a tool providing more flexibility of calculation PLATINUM provides a possibility to change some of these parameters F
30. r complexes PLATINUM numerically estimates hydrogen bonds stacking and hydrophobic intearactions in receptor ligand complexes This feature can be used to filter docking poses generated by jther software PLATINUM generates 2D hydrophobicity maps for certain molecular systems lipid bilayers and a helical peptides This feature included in version 1 1 facilitates analysis of hydrophobic hydrophilic interactions in complex biomolecular systems 1 Theory 1 1 Docking and Scoring PLATINUM does not perform ligand docking or any geometry optimization it only uses given 3D coordinates of biological molecules to calculate and visualize their hydrophobic hydrophilic properties PLATINUM also calculates hydrophobic hydrophilic match between two interacting molecules These are referred to as receptor and ligand However all this 1s relative user can supply a lipid bilayer as receptor and a peptide as the ligand or even vice versa Hydrophobic hydrophilic match can be used to filter the results of molecular docking This can be particularly useful in certain cases e g docking of nucleotides Pyrkov et al 2007 where hydrophobic interactions play the crucial role in molecular recognition and therefore deserve more precise analysis To facilitate the analysis of docking poses multiple ligands can be uploaded with the same structure of receptor 1 2 Molecular Hydrophobicity Potential MHP Hyrophobic effect plays crucial role in biom
31. t are closer than 4 to the point j are effectively contributing to the MHP MAP offset 1 5 1 5 Empirical atomic hydrophobicity constants may need some correction to produce more realistic picture of distribution of molecular properties We found that moderate shift to the more hydrophobic range improves the distribution of properties for low molecular weight compounds particularly nucleosides and nucleotides Fig 2 3 Meanwhile such moderate shift does not influence protein MHP greatly So it 1s recommended to use MHP offset of 0 03 logP units which is added to each atomic hydrophobicity constant prior to any MHP calculations User can change this value in the range 1 5 1 5 which covers the tabulated atomic constants To return to initial parameters as they appear in the parameterization table the MHP offset should be set to zero 14 MAP snin ligand 1 0 1 0 MAP snig receptor 1 0 1 0 These parameters are similar to the MHP offset but allow more flexible tuning of MHP values While the MHP offset adds the same constant to both the ligand and the receptor atoms MHP shifts provide separate change for these molecules The difference is that MHP shifts are added not to atoms but to MHP at surface points or grid points However the effect of MHP offset 0 5 and MHP shift 0 5 1s almost the same Both MHP shift values can be varied in the range 1 0 1 0 MHP offset 8 0
32. t be processed and is omitted however this warning will likely lead to unavoidable problems in MHP type assignment due to loss of atom s and finally to the program termination inspect the corresponding file manually ERROR bad file format program terminated a string corresponding to an atom does not contain enough information about this atom some data missing while in case of pdb or gro the atom is simply omitted PLATINUM is less forgiving in case of mol2 and sdf since these are more complex molecular formats ERROR bad file format wrong atom numbering in mol2 and sdf formats bonds are identified by the two numbers of corresponding atoms so atom numbering is quite important if some troubles with atom numbering were encountered program will terminate inspect the corresponding file manually WARNING atom type column not found trying to detect atom types automatically old pdb file format does not contain the atom element symbol in column 78 the new format does this is only a minor mistake since usually it 1s straightforward to identify the element type from the atom name for gro format it is the default procedure however be aware of difficulties that arise if atom name starts with characters other than its element type number 1s ok 1H will be identified as Hydrogen C8 will be identified as Carbon CL8 will be identified as Chlorine AC8 e g adenine C8 will be identified as a weird element A and omitted
33. ue However this should only be done with the optimal MHP parameters for rescoring results of GOLD dotdensity Low MHPoffset 0 03 MHPshift lig 0 6 MHpPshift rec 0 3 GOLD docking program was chosen as one of the most efficient and popular tool Kitchen et al 2004 In future we also plan to optimize scoring criteria for other popular docking programs Note that the numbers of h bonds and stacking are fractional This is because these types of intermolecular contact are described by weighting functions based on geometrical criteria Therefore only perfect geometry receives whole number equal to 1 0 For detailes on stacking see Section 2 3 H bond weighting function is described elswhere Pyrkov et al 2008 19 2 5 Visualization of Hydrophobic Hydrophilic Properties in Jmol When the results of docking have been processed and hydrophobic complementarity calculated one may wish to select some of them for more thorough study e g seek what part of ligand may be responsible for non optimal hydrophobic or hydrophilic contact or analyze the overall distribution of ligand hydrophobic properties For that purpose PLATINUM provides tools for visualization of hydrophobic hydrophilic properties and their complementarity between ligand and receptor molecules One such tool is visualization in Jmol applet http www jmol org on the PLATINUM web server To use it simply click the molecule icon just next to the ligand pose
34. y the value of TcIFP can vary between 0 different poses and 1 identical poses A weak point of the IFP comes from its definition Criteria to judge whether a contact is formed or not are not universal Thus although IFP was introduced only some years ago a number of different approaches to treat the interaction bits has already been described in literature This 1s because besides such contacts as hydrogen bond with well defined geometrical parameters there exist stacking or hydrophobic contacts for which different criteria may be applied For stacking this problem can be solved to some extent by data extraction from structural databases Pyrkov et al 2007 Meyer et al 2003 Meanwhile for hydrophobic interaction this task 1s quite more challenging since the nature of forces driving this interaction 1s still poorly understood So in PLATINUM we used our own approach to describe ligand receptor contacts These contacts and the rules defining their formation are listed in Table 1 1 However later these may undergo improvement and any remarks are welcome Table 1 1 Interaction bits in IFP Interaction bit comments pee or os Se A acceptor D donor H or 1N Case OF group i lt RDA 105 20 hydrogen R atom next to OH group h bond Same receptor is a protein molecule and its to protein main chain atom is named O or N in the structure file salt bridge N atom with formal negative charge
35. ydrophobic and hydrophilic properties between receptor and ligands we mean that hydrophobic hydrophobic and hydrophilic hydrophilic receptor ligand contacts are favorable or complementary and hydrophobic hydrophilic contacts are anticomplementary unfavorable PLATINUM calculates hydrophobic hydrophilic complementarity based on surface area of favorable and unfavorable contacts which are attributed to one of the six types Fig 1 3 LL Lipophilic ligand Lipophilic receptor LH Lipophilic ligand Hydrophilic receptor LH Lipophilic ligand Hydrophilic solvent water HL Hydrophilic ligand Lipophilic receptor HH Hydrophilic ligand Hydrophilic receptor HH Hydrophilic ligand Hydrophilic solvent water Figure 1 3 Types of contacts between ligand and its environment on ligand molecular surface Each dot type corresponds to different combination of ligand and environmental MHP i e hydrophobic hydrophilic properties Character L means lipophilic brown oil colored and H hydrophilic blue water colored Apostrophe symbol designates exposed ligand surface which is in contact with hydrophilic solvent Hereinafter each of these codes will be used to designate the corresponding surface area for example S is the area of ligand molecular surface corresponding to dots of type LL To rank the uploaded list of complexes the user may choose between different criteria See sect
36. you wish to view This immediately opens a new window where you can see three Jmol views of the same ligand pose in the receptor binding site The ligand surface is colored respectively by its own MHP left panel MHP induced by its environment middle panel and complementarity of hydrophobic hydrophilic properties between them right panel Ligand MHP Receptor or Water induced MHP Match nonpolar match polar mismatch Molecular Surface Transparency 50 O Show Other Molecule s as Wireframe v Molecular Surface Color Scheme blue white brown C Show Reference Figure 2 6 Visualization in online Jmol applet User can choose surface transparency and MHP color scheme to display the ligand own left panel and its receptor middle panel hydrophobic hydrophilic properties The right panel maps match grey and mismattch orange of the properties of the ligand and its environment 20 2 6 Saving Data to View MHP in a Molecular Visualization Application Another way to view the distribution of molecular hydrophobic hydrophilic properties is to save precalculated MHP data as files on a local disk These files may be then loaded into a molecular visualization application for analysis e g in PyMOL DeLano 2002 http www pymol org Also doing so you can create a nice picture or illustration of molecular hydrophobic properties and complementarity Furthermore such data may be useful to perform private cal
Download Pdf Manuals
Related Search