[gmx-users] Umbrella sampling sanity check

[Justin Lemkul]

On 8/10/15 2:27 AM, Nash, Anthony wrote:
> Hi all,
>
> I would appreciate a little sanity check for umbrella sampling pull code
> parameters.
>
> My reaction coordinate is defined as the distance between a globular
> soluble enzyme and a substrate in solution (water). I have already
> captured the individual windows using the trajectory generated from an
> earlier meta-dynamics simulation. I would now like to perform umbrella
> sampling, a technique which I feel is more accountable for the free
> energies it generates once a reaction coordinate is known.
>
> My two questions:
> 1) My earlier meta-dynamics simulation used the distance between the
> enzyme active site zinc ion to a particular substrate atom. The substrate
> atom (a backbone carbonyl oxygen) would in reality coordinate with the
> zinc prior to proteolysis. In pull code, can pull_group0 and pull_group1
> simply be the same two atoms (Zn2+ and O)? As the distance harmonic
> potential works on COM, this would just be a point of an atoms.
>

If both groups are those two atoms, by definition, your COM separation would 
always be zero and no potential would be applied.  You want the Zn2+ and O in 
separate groups, e.g.

pull_group0 = Zn2+
pull_group1 = O

> 2) Up until now I¹ve always worked on bilayer, so forgive the simplicity.
> Given that my substrate could diffuse in any direction, should I set
> pull_dim=Y Y Y?
>

Yes.

-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

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