[gmx-users] How much should epsilon-r be set?

[Justin Lemkul]

On 1/14/14, 8:07 AM, 李晴 wrote:
> Dear gmx-users,
> I use GMX to simulate a protein 700 peptide long with a ligand. The protein-ligand complex is put in a box with explicit water. I use AMBER99 force field.
> The epsilon-r was set with the default value 1, with coulombtype=PME. The problem is, the energy generated from .xtc file using g_energy showed large difference between coulomb energy and VDW potential energy. For example, I calculate the energy between the ligand and the protein, which were, Coul-SR=-406.172, LJ-SR=-115.504. We thought during the protein-ligand interaction, the coulomb energy and VDW energy should have similar values. But the results seemed to indicate that coulomb force had a much more important impact on the interaction.
> Is there any problem I set the epsilon-r=1? Should I change it larger to weaken the coulomb impact?

No.

http://gromacs.5086.x6.nabble.com/What-is-epsilon-r-td4888079.html

What makes you believe the van der Waals and Coulombic interactions should be 
equal in terms of energy?  If there is some experimental reason for that 
statement, then the results simply suggest that your parametrization of the 
ligand is incorrect.

-Justin

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

Justin A. Lemkul, Ph.D.
Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441

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