[gmx-users] Re: Recommended parameters for NVE simulation of SPCE water

[Andrew DeYoung]

Dear Professor Shirts,

Thank you so much. I really appreciate your time in responding to my post
and providing advice.  I will try your suggested parameters for reasonably
good energy conservation, and I will let you know how it goes.  

Thanks again for your time.  I will be in touch.

Andrew

------------------------------
Date: Wed, 10 Aug 2011 17:50:13 -0400
From: Michael Shirts <michael.shirts at virginia.edu>
Subject: Re: [gmx-users] Recommended parameters for NVE simulation of
	SPCE	water
> 1.  NOTE 1 above suggests that I use vdwtype = Shift.  When I do this, do
> you recommend that I apply long range dispersion corrections for both
energy
> and pressure, using DispCorr = EnerPres, or for only energy, using
DispCorr
> = Ener?  Typically, for various (non-NVE) calculations, I have been using
> DispCorr = no, but I am not sure if this is a good idea.  Pages 97-98 of
the
> Gromacs 4.5.4 manual seem to suggest that the energy correction due to
> DispCorr is small and usually only significant for free energy
calculations
> (which I will not be doing here).  As a rule of thumb, do you typically
turn
> dispersion corrections off?

For constant pressure simulations, or for reaching the constant
pressure equilibrium simulation, you should definitely include a
dispersion correction -- the density will be too large, and will be
cutoff dependent.

For constant volume simulations, the dispersion correction will be
constant.  It will thus NOT affect energy conservation, but WILL
affect average potential energy and average total energy,
significantly.

> 2.  NOTE 2 above suggests that I use either coulombtype = PME-Switch or
> coulombtype = Reaction-Field-zero.  Do you have any advice or
> recommendation?

For pretty good energy conservation, I would suggest:

rlist                           = 1.3
coulombtype              = PME
rcoulomb                    = 1.1
vdw-type                    = Switch
rvdw-switch                = 1.0
rvdw                          = 1.1

This should work quite well -- you might get some drift after 1-2 ns,
but not much.  I'm working on developing suggested PME parameters
right now for highly quantitative work, but it's not quite ready yet.


~~~~~~~~~~~~
Michael Shirts
Assistant Professor
Department of Chemical Engineering
University of Virginia
michael.shirts at virginia.edu
(434)-243-1821