[gmx-users] Protein and Ligand position restraint

[Justin Lemkul]

On 2/24/17 1:19 AM, Amir Zeb wrote:
> Hello gmx users,
>
> I want to conduct md simulation to explore the comparative stability and
> interaction mechanism of the reference compound (experimentally determined)
> and the newly screened candidate molecules by virtual screening approach.
> The gromacs version 5.0.7 is installed on my cluster. I want to listen to
> your suggestions regarding:
>
> 1. Is it rational to restraint the position of the protein and/or ligand?
> 2. If position restraint is necessary, which part of the system should I
> restraint?
> 3. Also, at which steps, I may restraint them, means NVT, NPT and md
> production? Because the Justin tutorial mentioned the release of restraint
> immediately after equilibration steps.
>
> I have co-factor FAD as the essential part of the protein, so I would like
> to consider protein-FAD as single moiety. I searched from gmx-user archive
> and Justin Tutorial, but I could not get a sound proof for this dilemma.
>
> Hope to listen to you soon
>

The purpose of position restraints is to allow the solvent to relax around the 
solute(s) of interest without perturbing them due to the initial artificiality 
in the system.  In this regard, it is sensible to restrain the protein and 
whatever bound ligand(s) may be present during equilibration.  Restraints during 
production make no sense.  Flexibility and dynamics are the point of running an 
MD simulation :)

-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

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