[gmx-users] parameters of em.mdp

[Justin A. Lemkul]

ahmet yıldırım wrote:
> Dear users,
> 
> I am using the Reaction-Field method for electrostatics interactions. I 
> used the following parameters for all input files (em.mdp, pr.mdp, 
> nvt.mdp, npt.mdp, md.mdp). I just changed as an epsilon_rf=78 in md.mdp. 
> If I set nstlist=rlist=rcoulomb=rvdw=1.0 for energy minimization, would 
> not it be better? What is your suggestions?
> 

Why do you think making such changes to the cutoffs would be better?  These 
settings, for the most part, are a fixed part of the force field you're using. 
Unless you have proof (either by your own demonstration or one that is 
published) that making such changes result in better results, you should avoid 
ad hoc changes.

> ; Neighbor Searching Parameters
> nstlist                 = 5 
> ns-type                 = Grid       
> pbc                = xyz           
> rlist                   = 0.9     
> ; Electrostatics
> coulombtype             = Reaction-Field 
> rcoulomb                = 1.4         
> epsilon_rf              = 54           
> ; VdW
> vdw-type                = Cut-off         
> rvdw                    = 1.4           
> *
> Another question:* I used 200 K (in pr.mdp) and 300 K (in nvt.mdp, 
> npt.mdp and md.mdp) the reference temperature for coupling. I analysed 
> the temperature after production run. I get "Temperature=312.646" 
> (g_energy -f md.edr -o temperature.xvg). that is, The temperature has 
> increased (approximately 12 K) during the simulation. What could be the 
> reason for the increase in temperature? I had setted to 200 K the 
> reference temperature for coupling in pr.mdp. it can cause?
> 

This outcome is precisely what you would expect, simply because you're using the 
reaction field method and it introduces cutoff artifacts.  Interestingly, this 
same outcome (an increase of exactly 12K) has been reported before:

http://lists.gromacs.org/pipermail/gmx-users/2009-January/039113.html

-Justin

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

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

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