[gmx-users] Rerun a period of trajectory with more frequent output.

[Justin Lemkul]

On 7/20/16 10:14 AM, Dd H wrote:
> Hi
> I have completed a protein-ligand MD simulation using Gromacs v4.5.5 and
> found protein undergoes a large conformational change. But sampling is too
> sparse to get intermediate states. In order to get some intermediate
> states I have to rerun the confromational change period of trajectory with
> more frequent coordinates output. So if there is a convenient way to do
> this? I don't want to rerun the whole trajectory since it's time-consuming.
>

If you have a checkpoint file from before the conformational change, you can 
supply it to grompp -t with an .mdp file that specifies more frequent output. 
This will be an exact continuation, but there is no guarantee that the 
trajectory will actually behave the same way as the first did due to the chaotic 
nature of MD.

If you don't have a checkpoint, you're out of luck aside from just doing the 
simulation over (which you should probably do anyway - one simulation does not 
generally prove anything).

-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

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