[gmx-users] How to calculate protein-surface interaction energy

[Justin Lemkul]

On 10/18/14 9:30 AM, James Lord wrote:
> Hi Justin,
>
> 1. Make sure your index groups are correctly constructed.
> Do I need to use make ndx? I thought just adding energygrps=protein surface
> at the end of .mdp is enough?
>

"Surface" isn't likely a default group (unless that's the name of the 
[moleculetype]) and grompp should have raised a fatal error if the group wasn't 
defined.

> 2. What does 5 nm correspond to?  A COM distance?  A minimum distance?  If
> the latter, then the LJ-SR energy should likely be zero, but if it's a COM
> distance, that's not necessarily the case.
> 5 nm is minimum distance
>

OK, in this case, the initial LJ-SR should be zero, provided that rvdw < 5.

> 3. Is your PE model united-atom or all-atom?  If the latter, you still have
> partial charges that will contribute to a negative Coul-SR (of course, this
> is something you can check very simply
> It is all atom

Then you must have partial charges on the individual atoms and therefore Coul-SR 
will be non-zero.

-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

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