[gmx-users] request

[Justin Lemkul]

On 2/25/13 6:09 AM, Group Gro wrote:
>
>
>
> Dear Gromacs Users.
>
>
> I apologize for asking a question that has come up several times by numerous users in the previous e-mails, but I have read the answers and suggestions to those posts and I am not still able to solve the problem based on them. It is completely possible that I am just not seeing the obvious, so I apologize, but I would welcome any advice. I want to work on "Human Glutathione Peroxidase" and downloaded its pdb file, but when I try pdb2gmx, I face Fatal Error:
>
> Program pdb2gmx, VERSION 4.5.5
> Source code file:
> /build/buildd/gromacs-4.5.5/src/kernel/pgutil.c, line: 91
> Fatal error:
> Atom CD is used in an interaction of
> type atom in the topology
> database, but an atom of that name was
> not found in residue
> number 50.
>
>
> I tried different force fields, for example OPLS-AA/L all-atom force field, different versions of AMBERs, but I
> always get the same problem! When I tried GROMOS's versions, the problem goes on CZ atom instead of CD. Here is a chunk of my pdb file (I have also attached the full pdb file to this e-mail):
>
> SHEET    6   A 1 PHE A 183  PRO A190  0
> SHEET    7   A 1 GLY A 192  HIS A200  0
> MODEL
> ATOM      1    N GLY A  38      -4.242  -9.833 -26.316  1.00  0.00           N
> ATOM      2   CA GLY A  38      -3.416 -10.949 -25.765  1.00  0.00           C
> ATOM      3    C GLY A  38      -1.953 -10.570 -25.632  1.00  0.00           C
> ATOM      4    O GLY A  38      -1.069
>   -11.266 -26.159  1.00  0.00           O
> ATOM      5    N THR A  39      -1.673  -9.461 -24.940  1.00  0.00           N
> ATOM      6   CA THR A  39      -0.286  -9.039 -24.705  1.00  0.00           C
> ATOM      7    C THR A  39      -0.073  -8.722 -23.214  1.00  0.00           C
> ATOM      8    O THR A  39      -1.035  -8.559 -22.442  1.00
>   0.00           O
> ATOM      9   CB THR A  39       0.124  -7.784 -25.533  1.00  0.00           C
> ATOM     10  OG1 THR A  39      -0.415  -6.602 -24.926  1.00  0.00           O
> ATOM     11  CG2 THR A  39      -0.309  -7.882 -27.037  1.00  0.00           C
> …..
> ATOM     91  OD1 ASP A  49      10.106   1.841  -5.471  1.00  0.00           O
> ATOM     92  OD2 ASP A  49      12.010   2.918  -5.273  1.00  0.00           O
> ATOM     93    N GLY A   50       8.192   0.495  -3.643  1.00  0.00           N
> ATOM     94   CA GLY A   50       6.805   0.919  -3.449  1.00  0.00           C
> ATOM     95    C GLY A   50       6.525   2.278  -4.054  1.00  0.00           C
> ATOM
>   96    O GLY A   50       5.449   2.818  -3.864  1.00  0.00           O
> ATOM     97    N GLU A   51       7.471   2.832  -4.802  1.00  0.00           N
> ATOM     98   CA GLU A   51       7.351   4.238  -5.211  1.00  0.00           C
> ATOM     99    C GLU A    51       6.737   4.438  -6.592  1.00  0.00           C
> ATOM    100    O GLU A
>     51       6.291   5.522  -6.897  1.00  0.00           O
> ATOM    101   CB GLU A  51       8.713   4.941  -5.174  1.00  0.00           C
> ATOM    102   CG GLU A  51       9.283   5.140  -3.759  1.00  0.00           C
> ATOM    103   CD GLU A  51       8.612   6.301  -3.006  1.00  0.00           C
> ATOM    104    N GLU A   52       6.729
>   3.418  -7.432  1.00  0.00           N
> ….
> ATOM    3035    OH TYR  A   228      -9.320
> -43.377 -52.577  1.00  0.00           O
> TER    3036      TYR A 228
> ENDMDL
> HETATM 3037   CL  CL 11000     -16.070
> -39.525  -6.168  1.00  0.00          CL
> HETATM 3038   NA  NA 11001      -1.679
> -34.065  -0.491  1.00  0.00          NA
> CONECT  996 1519
> CONECT 1021 1519
> CONECT 1519  996 1021
> END
>
> I know that the residue number printed
> by pdb2gmx is actually starting from zero, so the problem is actually in GLU.
> In both rtp file and pdb I have CD atom. Some suggestions wanted to
> change the order of C and O in pdb file, but it didn't work. What should I do?

Your statements are inconsistent.  You say above that changing the force field 
triggers a problem with a different atom (CZ instead of CD), which doesn't make 
much sense.  The amino acids have standard atom nomenclature, so changing the 
force field should not change the error.  Beyond that, the real reason is that 
you're missing an atom somewhere and the PDB header should have clearing 
"MISSING" lines that tell you exactly what that is.  If you need further help 
diagnosing, providing the actual PDB code would be helpful.

-Justin

-- 
========================================

Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================