[gmx-users] The output .Gro cannot be generated

[Mark Abraham]

Panichakorn Jaiyong wrote:
> Hi,

I think you would benefit from doing some tutorial material available on 
the GROMACS website.

> My molecule is the cycloheptapeptide which contains a proline ring. The 
> first  problem is the disapperance of 7 hydrogen atoms in rtp file, so I 
> run pdb2gmx with -ignh to ignore them and using Gromacs force field. The 
> second one is the warning that atom CA not found in residue 8 while 
> adding atom.

7 atoms are not disappearing from an .rtp file when you use pdb2gmx. Why 
are you ignoring the advice that pdb2gmx gives that the GROMACS force 
field is deprecated?

I am not aware that pdb2gmx treats cyclic structures. In any case, you 
should start with a linear version of your peptide and get that working 
correctly first.

To actually get a cyclic .top working, you will need to tweak the 
topology by hand to close the final peptide bond yourself. You will need 
to read chapter 5 of the manual thoroughly, and work by analogy from the 
other peptide bonds in your .top file.

> How does to solve with this?
> I try to pose the pdb in Prodrg site for a request of the .top file, but 
> anything turns to me for 3 weeks.

It seems likely you made some mistake, like the wrong email address, or 
not submitting.

> It appears after select a forcefield like this:
> 
> Opening library file /usr/share/gromacs/top/ffgmx.rtp
> Opening library file /usr/share/gromacs/top/aminoacids.dat
> Reading TrialcordyhepA.pdb...
> ', 64 atomshepA
> Opening library file /usr/share/gromacs/top/xlateat.dat
> 26 out of 26 lines of xlateat.dat converted succesfully
> Analyzing pdb file
> There are 1 chains and 0 blocks of water and 22 residues with 64 atoms
> 
>   chain  #res #atoms
>   1 ' '    22     64
> 
> All occupancies are one
> Opening library file /usr/share/gromacs/top/ffgmx.atp
> Atomtype 54
> Reading residue database... (ffgmx)
> Opening library file /usr/share/gromacs/top/ffgmx.rtp
> Residue 98
> Sorting it all out...
> Opening library file /usr/share/gromacs/top/ffgmx.hdb
> Opening library file /usr/share/gromacs/top/ffgmx-n.tdb
> Opening library file /usr/share/gromacs/top/ffgmx-c.tdb
> Processing chain 1 (64 atoms, 22 residues)
> There are 7 donors and 8 acceptors
> There are 12 hydrogen bonds
> Checking for duplicate atoms....
> Now there are 23 atoms. Deleted 41 duplicates.
> Opening library file /usr/share/gromacs/top/specbond.dat
> 5 out of 5 lines of specbond.dat converted succesfully
> N-terminus: NH3+
> C-terminus: COO-
> WARNING: atom CA not found in residue 8 while adding atom
> -------------------------------------------------------
> Program pdb2gmx, VERSION 3.3
> Source code file: genhydro.c, line: 304
> 
> Fatal error:
> Atom CA not found in residue Tyr8 while adding hydrogens
> -------------------------------------------------------

If residue 8 doesn't exist, then pdb2gmx doesn't handle cyclic peptides.

Mark