Chimera vs Amber, which one digresses from the average pdb rules?
A lipid/water pdb file that was good for Leap (Amber9) to get top/crd, once passed through Chimera 2470 to remove residues is no more good for Leap. All TER were removed by Chimera, ATOM renamed HETATM, while CONECT records were added. As such, Leap said: In file [chirality.c], line 142 Atom named C217 from POP did not match Aborting I knew such type of message. There is no C217 atom. On previous (different) occasions, insertion of TER records put the house in order. I inserted all TERs between POPC residues and between WAT residues (thanks to Python!). Now Leap said: In file [atom.c], line 444 bond AtomProblem found Aborting As I did not know this type of message, I tried to investigate that line 444. Following vim's command :n,444 that line, indicated at bottom of terminal as "444,1", reads HETATM 441 O31 POPC 5 -44.082 -12.831 14.685 1.00 0.00 O A carbon atom (if "[atom.c]" means carbon, is at line 447,1, which reads HETATM 444 C32 POPC 5 -46.343 -12.925 15.140 1.00 0.00 C In the original (good for Leap) pdb those lines read: ATOM 441 O31 POPC 4 -37.722 -19.453 18.019 1.00 0.00 L11 O ATOM 444 C32 POPC 4 -39.844 -19.273 18.941 1.00 0.00 L11 C Chimera also changed atom names, such as in the final part of the POPC residue (Sybyl naming?, if so, there is an alternative in Chimera not to follow Sybyl rules?) HETATM 528 XH14 POPC 5 -47.646 -9.424 -0.112 1.00 0.00 H HETATM 529 YH14 POPC 5 -46.217 -10.553 -0.172 1.00 0.00 H HETATM 530 5C31 POPC 5 -47.890 -11.168 -1.419 1.00 0.00 C HETATM 531 XH15 POPC 5 -47.763 -12.309 -1.296 1.00 0.00 H HETATM 532 YH15 POPC 5 -48.986 -11.063 -1.417 1.00 0.00 H HETATM 533 6C31 POPC 5 -47.242 -10.566 -2.756 1.00 0.00 C HETATM 534 XH16 POPC 5 -46.135 -10.759 -2.831 1.00 0.00 H HETATM 535 YH16 POPC 5 -47.802 -11.112 -3.597 1.00 0.00 H HETATM 536 ZH16 POPC 5 -47.451 -9.454 -2.737 1.00 0.00 H TER which, in the original (good) pdb was: ATOM 528 H14X POPC 4 -38.665 -14.879 4.398 1.00 0.00 L11 H ATOM 529 H14Y POPC 4 -39.080 -16.603 3.952 1.00 0.00 L11 H ATOM 530 C315 POPC 4 -40.370 -15.221 2.996 1.00 0.00 L11 C ATOM 531 H15X POPC 4 -41.330 -15.752 3.099 1.00 0.00 L11 H ATOM 532 H15Y POPC 4 -40.694 -14.148 3.019 1.00 0.00 L11 H ATOM 533 C316 POPC 4 -39.787 -15.573 1.664 1.00 0.00 L11 C ATOM 534 H16X POPC 4 -39.553 -16.661 1.544 1.00 0.00 L11 H ATOM 535 H16Y POPC 4 -40.534 -15.211 0.948 1.00 0.00 L11 H ATOM 536 H16Z POPC 4 -38.782 -15.094 1.533 1.00 0.00 L11 H TER Although it is not clear to me yet where really Leap crashed (nor am I placing Leap as the standard; simply I have to get top/crd from it, if I want to carry out MD with Amber). Elaine already kindly clarified the ATOM/HETATM issue. I hope now to get help about the atom names , ie a table of conversion to get Leap working or a not-Sybyl choice in Chimera. Once again, I regret very much to pose so many problems of language. Thanks francesco pietra ____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs
Hi Francesco, In Chimera we have worked to create PDB files in the correct format, as described at the PDB web site: http://www.wwpdb.org/docs.html We also have a brief intro to this format in the Chimera documentation: http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/tutorials/ framepdbintro.html We decided the best approach would be to follow the format standard, including recent changes. However, lots of other programs don't follow the standard, and when using multiple programs it is usually necessary to hand-edit or use scripts to convert files. In version 2.3 PDB format, the atom name occupies columns 13-16 with the element symbol right-justified in 13-14. So if the element symbol is one character, it is supposed to be in column 14, not 13, i.e. C315 should be 5C31. The output you are getting follows this role. In version 3.0 PDB format, that is still true, except for hydrogen atoms. Both versions follow the HETATM/ATOM rules I mentioned earlier. Best, Elaine ----- Elaine C. Meng, Ph.D. meng@cgl.ucsf.edu UCSF Computer Graphics Lab and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html
--- Elaine Meng <meng@cgl.ucsf.edu> wrote:
Hi Francesco, In Chimera we have worked to create PDB files in the correct format, as described at the PDB web site: http://www.wwpdb.org/docs.html
We also have a brief intro to this format in the Chimera documentation: http://www.cgl.ucsf.edu/chimera/docs/UsersGuide/tutorials/ framepdbintro.html
We decided the best approach would be to follow the format standard, including recent changes. However, lots of other programs don't follow the standard, and when using multiple programs it is usually necessary to hand-edit or use scripts to convert files.
In version 2.3 PDB format, the atom name occupies columns 13-16 with the element symbol right-justified in 13-14. So if the element symbol is one character, it is supposed to be in column 14, not 13, i.e. C315 should be 5C31. The output you are getting follows this role.
In version 3.0 PDB format, that is still true, except for hydrogen atoms.
Here I really don't understand. With your "including recent changes" above, I understand that you follow 3.0 or 3.1 PDB format. Unless you add that 3.0 and 3.1 formats are exercises not to follow (the official pdb booklet I have, is V. 2.3, added of V. 3.0 and V 3.1). If you follow 3.0, modification of Amber's file ATOM 531 H15X POPC 4 -41.330 -15.752 3.099 1.00 0.00 L11 H ATOM 532 H15Y POPC 4 -40.694 -14.148 3.019 1.00 0.00 L11 H ATOM 533 C316 POPC 4 -39.787 -15.573 1.664 1.00 0.00 L11 C ATOM 534 H16X POPC 4 -39.553 -16.661 1.544 1.00 0.00 L11 H ATOM 535 H16Y POPC 4 -40.534 -15.211 0.948 1.00 0.00 L11 H ATOM 536 H16Z POPC 4 -38.782 -15.094 1.533 1.00 0.00 L11 H TER by Chimera, as follows (as it happened to me, see previous mail): HETATM 129 XH15 POPC 1 -40.149 -41.259 3.406 1.00 0.00 H HETATM 130 YH15 POPC 1 -39.107 -42.686 3.702 1.00 0.00 H HETATM 131 6C31 POPC 1 -40.650 -42.850 2.121 1.00 0.00 C HETATM 132 XH16 POPC 1 -41.411 -42.165 1.778 1.00 0.00 H HETATM 133 YH16 POPC 1 -39.859 -42.645 1.328 1.00 0.00 H HETATM 134 ZH16 POPC 1 -40.899 -43.957 2.212 1.00 0.00 H breaks the H rules. Finally, everything in the "classic" are is in view of MD. Even DOCK is recommended as a preliminary to MD (by both Scott Brozell and D case). Therefore, it is a pity - to say politely - having to spend so much effort in rectifying files from pre-programs. Unless Chimera is bound to some other MD suite than Amber, to which MD suite Chimera's files flow in without barriers. This is under my sincere assertion that I consider Chimera a gem. The gift of a gem. And it is far from me to try to engage in discussion on matter not of my competence, like these ones. My only aim is: once TERs are inserted, as I did, I have "simply" to recondition the files for the position of single-character elements (including H). This is what I'll try, allowing for an European night of break to see if contrary suggestions come out. Lot of users should have had similar problems. Thanks a lot Francesco
Both versions follow the HETATM/ATOM rules I mentioned earlier. Best, Elaine ----- Elaine C. Meng, Ph.D. meng@cgl.ucsf.edu UCSF Computer Graphics Lab and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html
____________________________________________________________________________________ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs
Hi Francesco, Yes, I saw the lines in your mail and have reported the hydrogen name issue as a bug. We intend to completely follow the PDB format. However, fixing the hydrogen naming will not affect the issues you were having with HETATM/TER and names like 5C13, because those are already handled correctly by Chimera (according to the PDB format standard). Format issues are not limited to Chimera users, of course, but anyone who works with multiple programs! Elaine ----- Elaine C. Meng, Ph.D. meng@cgl.ucsf.edu UCSF Computer Graphics Lab and Babbitt Lab Department of Pharmaceutical Chemistry University of California, San Francisco http://www.cgl.ucsf.edu/home/meng/index.html
On Nov 23, 2007, at 12:56 PM, Francesco Pietra wrote:
Here I really don't understand. With your "including recent changes" above, I understand that you follow 3.0 or 3.1 PDB format. Unless you add that 3.0 and 3.1 formats are exercises not to follow (the official pdb booklet I have, is V. 2.3, added of V. 3.0 and V 3.1). If you follow 3.0, modification of Amber's file
As Elaine mentioned, v2.3 PDB files would place 'H' in column 14, so a 4-letter hydrogen name would be something like 1HD1, whereas in v3.0 they decided to just have the H start the name, producing names like HD11. The fact that Chimera still produces v2.3 hydrogen names is wrong, and I have added your name to the bug-database entry for producing 3.0 names, but nonetheless the PDB changeover to 3.0 only happened a few months back, so most programs should be able to handle both forms.
Finally, everything in the "classic" are is in view of MD. Even DOCK is recommended as a preliminary to MD (by both Scott Brozell and D case). Therefore, it is a pity - to say politely - having to spend so much effort in rectifying files from pre-programs. Unless Chimera is bound to some other MD suite than Amber, to which MD suite Chimera's files flow in without barriers.
Chimera isn't "bound" to any MD suite in particular. It simply tries to produce standard PDB files. The only software we've made a conscious attempt to coordinate with is DOCK. --Eric Eric Pettersen UCSF Computer Graphics Lab
Eric: Thanks. Please see below. --- Eric Pettersen <pett@cgl.ucsf.edu> wrote:
On Nov 23, 2007, at 12:56 PM, Francesco Pietra wrote:
Here I really don't understand. With your "including recent changes" above, I understand that you follow 3.0 or 3.1 PDB format. Unless you add that 3.0 and 3.1 formats are exercises not to follow (the official pdb booklet I have, is V. 2.3, added of V. 3.0 and V 3.1). If you follow 3.0, modification of Amber's file
As Elaine mentioned, v2.3 PDB files would place 'H' in column 14, so a 4-letter hydrogen name would be something like 1HD1, whereas in v3.0 they decided to just have the H start the name, producing names like HD11. The fact that Chimera still produces v2.3 hydrogen names is wrong, and I have added your name to the bug-database entry for producing 3.0 names, but nonetheless the PDB changeover to 3.0 only happened a few months back, so most programs should be able to handle both forms.
I succeed in letting LEaP (Amber9) reading pdb files written by Chimera. Which may still be in the economy in view of "delete overlapping residues" possibility in Chimera.
Finally, everything in the "classic" are is in view of MD. Even DOCK is recommended as a preliminary to MD (by both Scott Brozell and D case). Therefore, it is a pity - to say politely - having to spend so much effort in rectifying files from pre-programs. Unless Chimera is bound to some other MD suite than Amber, to which MD suite Chimera's files flow in without barriers.
Chimera isn't "bound" to any MD suite in particular. It simply tries to produce standard PDB files. The only software we've made a conscious attempt to coordinate with is DOCK.
Yes, and that is an extremely useful coordination. However, DOCK is strictly bound to Amber. I don't simply mean the DOCK's built-in amber_rescore, but Scott Brozell indication to run Amber's MD after the series of dockings. I can say again that in my opinion Chimera is a gem. I could not have done what I have done so far in docking-MD without Chimera. Now - with side help by VMD - I have the protein-ligand complex in a membrane, the whole solvated. Though, a choice of different protocols when saving pdb files from Chimera ("saving for LEaP) would speed up the work by certain users. Is it a major modification of software? Cheers francesco
--Eric
Eric Pettersen UCSF Computer Graphics Lab
_______________________________________________ Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://www.cgl.ucsf.edu/mailman/listinfo/chimera-users
____________________________________________________________________________________ Be a better sports nut! Let your teams follow you with Yahoo Mobile. Try it now. http://mobile.yahoo.com/sports;_ylt=At9_qDKvtAbMuh1G1SQtBI7ntAcJ
Hi Francesco, According to the AMBER mailing list, there is a program that comes with AMBER that changes hydrogen names from standard PDB to "AMBER PDB." See these links: http://amber.ch.ic.ac.uk/archive/200704/0377.html http://amber.scripps.edu/doc6/html/AMBER-sh-17.3.html I do not know if it changes the names of non-hydrogen atoms or works with version 3.0. You may want to ask the AMBER people about this, or if there are any newer versions of this programs or similar scripts. You may also want to take a look at Simulaid (free for academics): http://atlas.physbio.mssm.edu/~mezei/simulaid/ It performs many simulation-related tasks including conversions of residue and atom names between various conventions including PDB and AMBER. Again, I have not tried it myself. Best, Elaine
participants (3)
-
Elaine Meng -
Eric Pettersen -
Francesco Pietra