[gmx-users] modified atomtypes.atp and grompp

Justin A. Lemkul jalemkul at vt.edu
Fri Mar 11 13:07:11 CET 2011

Rausch, Felix wrote:
> Dear Gromacs users,
> I would like so simulate a protein with phosphorylated amino acids with 
> Gromacs 4.5.3. To do so, I got the "ffG43a1p" forcefield available from 
> the user contributions section (subsection force fields) and adapted the 
> parameters given there to Gromos53a6.
> The problem now is that they introduced a new atom type called "OP", 
> which I have to add to the atomtypes.atp of my modified force field. I 
> already read about problems of a modified atomtypes.atp here in the 
> mailing list, but I assumed the problems were sorted out with version 4.5.3.

Not entirely, but probably the post you saw related to an LP atomtype, which 
appears to be a special case.

> But if I add OP to the atom list, pdb2gmx works fine, but grompp is 
> raising the error
> Program grompp, VERSION 4.5.3.
> Source code file: toppush.c, line: 629
> Fatal error:
> Unknown atomtype CMET
> which is quite strange because  "CMET" isn't appearing in any file I'm 
> using. When I remove OP from the atom list, pdb2gmx is of course raising 
> an error, so I'm a bit stuck here...

grompp reads all the force field files and is finding something it doesn't like. 
  In 4.5.3, the proper capitalization should be "CMet," not "CMET," as it was in 
old versions.  Capitalization and other nomenclature has been standardized as of 
version 4.5.3, so old files with outdated names cause errors.


> To cut a long story short: Is there any way to introduce a new atom type 
> to atomtypes.atp or do I have to parameterize the phosphorylated 
> residues with already presend atom types?  
> Thank you in advance,
> Felix Rausch


Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080


More information about the gromacs.org_gmx-users mailing list