[gmx-users] Pdb2gmx melts molecules due to bad distances

[Justin Lemkul]

On 11/27/17 6:04 AM, GIANMARCO BARTALINI wrote:
> Hello guys, I produced a system "protein + lipid membrane" using the online
> tool CHARMM-GUI. Since I need to use the Amber force field, I added the
> Lipids17 force field inside Gromacs. The implementation should work fine,
> since the conversion of an equilibrated structure produces a nice and
> correct gro file. But when I try to convert the (not minimized) PDB file
> that I get from CHARMM-GUI using pdb2gmx, I get a gro structure where many
> lipid molecules are melted together (due to their bad distances). If I use
> this structure for a minimization, the calculation stops due to bad
> energies. Do you know how to solve this issue?

This isn't a pdb2gmx problem so much as it is a quirk of how CHARMM-GUI 
creates lipid membranes. There is almost always overlap between tail 
atoms in the different lipid leaflets. The CHARMM minimizer deals with 
these just fine, but GROMACS almost always chokes. The reason is that 
GROMACS sets a rather aggressive default EM step size of 0.02 nm, 
whereas the CHARMM minimizer defaults to 0.002. Lower the value of 
emstep and see if it proceeds correctly. That has always worked for us.

-Justin

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

Justin A. Lemkul, Ph.D.
Assistant Professor
Virginia Tech Department of Biochemistry

303 Engel Hall
340 West Campus Dr.
Blacksburg, VA 24061

jalemkul at vt.edu | (540) 231-3129
http://www.biochem.vt.edu/people/faculty/JustinLemkul.html

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