AMBER Archive (2005)Subject: Re: AMBER: collision frequency in GBSA dynamics
From: David A. Case (case_at_scripps.edu)
Date: Tue Jan 25 2005  16:23:34 CST
On Tue, Jan 25, 2005, Giulio Rastelli wrote:
>
> I am running a molecular dynamics calculation on a protein structure using GBSA
> (igb=5, gbsa=1) using amber8. Using Langevin dynamics (ntt=3) one need to se
> the collision frequency (gamma_ln) to something different from the default of
> zero...what value of gamma_ln would be "appropriate" for this gb calculation?
The friction term for water is close to 50 ps^1, but it is generally
advantageous to use a much smaller value, say 5 10. You should be aware that
no value really is a good approximation to the *dyanmics* of a water system,
but values around 5 generally do a good job of sampling the configurations
one expects for a solvated room temperature simulation.
Take a look at this paper:
%A R.J. Loncharich
%A B.R. Brooks
%A R.W. Pastor
%T Langevin dynamics of peptides: The frictional dependence of isomerization
rates of NactylananylN'methylamide
%J Biopolymers
%V 32
%P 523535
%D 1992
...good luck...dac

The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
