[gmx-users] Minimisation failure: The charge group moved more than the distance allowed by the domain decomposition

[Justin Lemkul]

On 7/4/14, 5:54 AM, Vytautas Rakeviius wrote:
> Here all my files:
>
> https://www.dropbox.com/s/94u9b9stsk84ckz/1oyn6wat1994-15a_vconf-docking_1.tar.bz2
>
>
> I try minimize this with:
>
> grompp_mpi_d -maxwarn 2 -f ~/emamber.mdp -c *_solvated_bonded_GMX.gro -p *_solvated_bonded_GMX.top -o em.tpr
> mdrun_mpi_d -deffnm em
>
>
> Here is my emamber.mdp file: http://pastebin.com/vtiuCzt0
> Here is my full error: http://pastebin.com/zX57DnQf
>
> I would be thankful for help.
>

Either the starting coordinates contain bad clashes or the topology is unstable. 
  The initial force is not insanely high, but often CG doesn't work well as the 
initial minimization.  Start with steepest descents and then do CG after steep 
has converged reasonably.  Otherwise, also note that mdrun tells you the exact 
atom number that's having a problem; start by visualizing what's going on in its 
surroundings.

-Justin

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

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA 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
http://mackerell.umaryland.edu/~jalemkul

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