[gmx-users] LIE method

[Justin Lemkul]

On 1/12/14, 5:18 PM, Mahboobeh Eslami wrote:
> dear Justin thanks a lot
> I use Amber03 force field and TIP3P water model for my system. following options
> are used in first my mdp file:
> nstlist  = 5
> rlist  = 1.0
> rcoulomb = 1.0
> rvdw  = 1.0
> I have following questions. if possible, please answer them.
> 1) are these options good for LIE method? do LIE method need especial options?

Run settings like these are dependent upon the force field.  LIE is simply a 
post-processing technique.

> 2) I saw following advice in gmx forum " save a lot of frames to get reasonably
> accurate numbers " (http://gromacs.5086.x6.nabble.com/PME-td4450969.html)
> do you have especial comment for frame saving?

No insight there.  I am not familiar with the details of what was proposed in 
that post.  As far as I know, most people do not attempt to re-figure energies 
like that; they simply re-process using RF or long cutoffs (RF is better).

> 3)  I use *coulombtype = *cutoff  and rcoulomb = 1.0 in new mdp file for
> rerun energies. are these options good or not?

No, plain cutoffs are very inaccurate.

-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

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