[gmx-users] protein-protein complex gets "separated" in EM step

[Justin Lemkul]

On 10/25/16 10:16 AM, Gregor Hagelüken wrote:
> Dear all,
>
> I am attempting to do a MD simulation off a protein-protein complex. I have successfully performed MD runs on the individual proteins.
> Now, I have taken the complex structure (X-ray structure) and ran it through pdb2gmx to get a topology. This also worked. The program does not complain and the complex structure in the resulting .gro file looks just like the input .pdb file in PyMOL.
>
> Next, I performed an energy minimization using these commands:
>
> grompp -v -f minim.mdp -c protein.gro -p protein.top -o protein-EM-vacuum.tpr
> mdrun -v -deffnm protein-EM-vacuum -c protein-EM-vacuum.gro
>
> This runs fine, but I noticed that in the resulting protein-EM-vacuum.gro file, the two components of my complex (chains A and B) are separated by ~50 Å. Even if I only run the minimisation for 5 steps.
> Any idea why this happens?
>
> Thank you very much for your help!
> Cheers,
> Gregor
>
>
>
> P.S.: I took the minim.mdp file from this website: http://cgmartini.nl/%7Emdcourse/md.html <http://cgmartini.nl/~mdcourse/md.html>
>

That .mdp file uses PBC in all dimensions, which is not normal for a vacuum 
minimization or simulation.  Your box is probably too small relative to the 
protein complex, and then the structures start to jump across PBC because you're 
using "pbc = xyz."  For a true in vacuo minimization or simulation, use "pbc = 
no" and the box size becomes irrelevant.

-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

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