AMBER Archive (2003)

Subject: Re: GB vs. explicit solvent calculations

From: David A. Case (case_at_scripps.edu)
Date: Thu Mar 20 2003 - 09:46:53 CST

On Thu, Mar 20, 2003, Thomas Steinbrecher wrote:
>
> >From my output files I see that 1ps of equilibration takes
> 1300 sec. with explicit solvent (cutoff 12A), while it
> takes 2700 sec. using the GB model, both calculated on a
> dual processor PC.

This is about what I would expect. The current version of sander is not
as efficient as one would like for larger proteins and these large cutoffs.
Your PME simulation will probably be even faster (and really just as accurate)
if you use cut=8 or 9 rather than 12. You can speed up the GB by setting
nrespa=4, dt=0.001 -- test this out, but it seems to work pretty well on
our systems. Depending on your accuracy/stability needs, you could try a
shorter cutoff (15-20 Ang.) in GB: remember that with saltcon turned on,
the potential dies off as exp(-kappa*r)/(epsilon*r), so that it is less
long-range than you may be used to for bare Coulomb interactions.

The next release will have a much faster GB implementation.

..hope this helps...dac 

-- 

==================================================================
David A. Case                     |  e-mail:      case_at_scripps.edu
Dept. of Molecular Biology, TPC15 |  fax:          +1-858-784-8896
The Scripps Research Institute    |  phone:        +1-858-784-9768
10550 N. Torrey Pines Rd.         |  home page:                   
La Jolla CA 92037  USA            |    http://www.scripps.edu/case
==================================================================