[gmx-users] Multiple chains, cofactors and ligands

YOLANDA SMALL yas102 at psu.edu
Thu Feb 2 18:12:03 CET 2006


Dear users,

I am having a problem with grompp which gives me the following message:
Fatal error: Invalid order for directive defaults, file
""/gromacs-3.2.1/share/top/ffgmx.itp"", line 4

My simulation details:
An enzyme with 2 chains, 2 cofactors and 2 ligands in a 115 Angstrom box of SPC
waters. The total system size is 150872 atoms.  Each chain has a separate .itp
file created with pdb2gmx. Each cofactor and ligand has an .itp file created
with PRODRG.  The files are joined in the following topology file:

; Include forcefield parameters
#include "ffgmx.itp"

; Include chain topologies
#include "chain_A.itp"
#include "chain_B.itp"
#include "cofactor_A.itp"
#include "cofactor_B.itp"
#include "ligand_A.itp"
#include "ligand_B.itp"
#include "Na.itp"

; Include water topology
#include "spc.itp"

; Include generic topology for ions
#include "ions.itp"

[ system ]
; Name
2 COMPLEXED WITH (Ligand)

[ molecules ]
; Compound        #mols
Protein_A           1
Protein_B           1
SOL               346
COF_A               1
COF_B               1
LIG_A               1
LIG_B               1
NA                  1
SOL             47872

The problem:
I only get this message when the topology of the cofactor and ligand are
included.  When I run the simulation on the protein chains only without the
ligand and cofactor, it runs just fine.  I double-checked the topology of the
ligand and cofactor and I don't see any errors. Do I need to change the
defaults in the following ffgmx.itp file to get this to work???? (The error
message points to line 4 of this file)  If so, what should I change the
defaults to?

#define _FF_GROMACS
#define _FF_GROMACS1

[ defaults ]
; nbfunc        comb-rule       gen-pairs       fudgeLJ fudgeQQ
  1             1               no              1.0     1.0

#include "ffgmxnb.itp"
#include "ffgmxbon.itp"

Thanks in advance for your help,
YAS



More information about the gromacs.org_gmx-users mailing list