[gmx-users] moleculaes breaking off after EM - help

Justin Lemkul jalemkul at vt.edu
Mon Oct 3 13:36:04 CEST 2016



On 10/2/16 10:09 PM, Thejus Kartha wrote:
> Greetings!
> I was performing an EM run with an array of dioxane molecules, for which I had produced co-ordinates from PRODRG. I got the co-ordinates, and made an array of it, using the gmx genconf module, and corrected my topology file. Then, I thought I should minimize the energy of this array before solvating it, or anything. So, as I was running the grompp module using mdp files from Dr. Lemkul's tutorials, it gave me the following error:
> Steepest Descents converged to machine precision in 641 steps,
> but did not reach the requested Fmax < 1000.
> Potential Energy  =  2.6059185e+09
> Maximum force     =  8.6477500e+04 on atom 557
> Norm of force     =  7.7685435e+03
>
> I am presuming there's something seriously wrong with my input files, as I've been told many times before. That said, how do I tweak around with my mdp file in this case? I mean, there's this Fmax value that I have to deal with. How do I know if I am using the right threshold for Fmax? Do I base this value on information from literature? And of course, when I open the .gro file after the EM, I find my molecules to be smashed to smithereens. There are points and broken triangles flying around in all directions (dioxane is a hexagon, and it has broken to half.) Why does this happen? Please help me.

This suggests your topology is wrong.  Test with one molecule in vacuo.  If it 
does the same thing, fix the topology.  If the in vacuo test goes well, then 
your coordinates in the assembled box are unreasonable.

-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

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


More information about the gromacs.org_gmx-users mailing list