[gmx-users] Generating topology for a HEM protein with CHARMM27

[Justin Lemkul]

On 5/12/16 10:27 AM, zeineb SI CHAIB wrote:
> Dear gmx users,
>
> I'm working on simulating a homodimer with its cofactors (hem, THG) and its substrate with GROMACS.
>
> Since my system is complex, I decided to start first with a monomere taking only protein residues and the hem group.
>
> When I run pdb2gmx, using CHARMM27 FF, I had this error:
> "Fatal error: There were 30 missing atoms in molecule Other_chain_A, if you want to use this incomplete topology anyhow, use the option -missing"!
>
> I think that this error is about hydrogen atoms but since my PDB is generated by X-RAY diffraction there is no hydrogen! What should I so?
>

Probably they are missing H atoms in heme, and there is no .hdb entry for heme.

Note that dealing with heme-bound proteins is difficult because the 
autogeneration of angles and dihedrals done by pdb2gmx actually assigns 
dihedrals that should not exist and you have to manually edit the topology file. 
  It's not pretty.  For complicated systems like this, I suggest using 
CHARMM-GUI, which uses CHARMM itself to correctly generate the topology and all 
the necessary GROMACS inputs.  It's quite simple.

-Justin

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

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

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