[gmx-users] gmx insert-molecules

[Justin Lemkul]

On 7/11/16 8:15 AM, Dave Michael wrote:
> Hi guys,
>
> I've got a system of 400 water molecules (
> http://md.chem.rug.nl/index.php/example-applications2/solvent-systems) and
> I want to make and simulate a system with 1k, 10k, 100k and 1M particles.
>
> So far, I've done:
> gmx insert-molecules -ci water.gro -o water.gro -box 4.59151 -nmol 400
> -radius 0.05
> (what I've tried doing is double the box volume and double the number of
> particles and it didn't work without -radius 0.05 / default radius doesn't
> add new molecules to the system)

Probably because insert-molecules doesn't understand whatever the particle type 
is that it's being provided unless you tell it.  CG types are often not anything 
like normal atoms.

> And I got a file with 800 particles, some of them have negative coordinates
> (I presume those are initial coordinates) and no initial velocities.
>

There are no velocities because insert-molecules doesn't serve that function, 
nor would any velocities be meaningful or useful in any way at this stage.

> Running (GROMACS 5.1.2) md on that file doesn't work as the simulation
> hangs @ step 0. Using gmx energy I can see that the total energy and
> temperature are off the scales but the density is fine (990 kg/m^3)
>

This suggests something is topologically incorrect.  Have you done energy 
minimization?

-Justin

> I don't have much experience in the chemical / physical aspects, can you
> please tell me (step by step if possible) what I missed and what should I
> do next?
>
> Dave
>

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

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

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