[gmx-users] coenzyme-protein complex

Justin Lemkul jalemkul at vt.edu
Sat Feb 1 01:48:53 CET 2014

On 1/31/14, 11:53 AM, Mostafa Javaheri wrote:
> Dear Justin
> I have a problem with centering the hetrodimer protein in the dodecahedron or
> octahedron box, the related commands are:
> 1.pdb2gmx -f A1CBIII-W3.pdb -ff charmm27 -water tip3p -ignh -o conf.pdb
> -nochargegrp -merge all -posrefc 1000 -renum
> 2.editconf -f conf.pdb -o boxed.pdb -c -d 1.0 -bt dodecahedron
> 3.genbox -cs -cp boxed.pdb -o solvated.pdb -p topol.top
> in the solvated.pdb output file protein represents at the corner of the box and
> some of it is out of box although it will not happen for -bt cubic (whole
> protein will be centered in the box), considering periodic boundary condition,
> mdrun will be ok; after five ns mdrun one of the protein's chain represents
> inside the box and the other chain out of the box in md.gro file. After running
> trjcov several times with different options (including -pbc mol, atom, nojump,
> whole -center, -ur compact) none of them could put the whole protein in one unit
> cell. I would be grateful again for your help.

Without seeing exactly what you've tried, it's rather futile to try to guess 
what to suggest.  Complexes are difficult to deal with.  The first step should 
almost always be trjconv -pbc nojump, but further iterations may vary.  See 
and also be aware that you can (and should, in many cases) use custom index 
groups for centering or other fitting, i.e. some residues at the protein-protein 
interface or something else that makes sense.  Centering on "protein" normally 
fails in such cases, for reasons discussed repeatedly in the list archive.



Justin A. Lemkul, Ph.D.
Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441


More information about the gromacs.org_gmx-users mailing list