AMBER Archive (2002)

Subject: using belly with GB

From: David A. Case (case_at_scripps.edu)
Date: Wed Aug 07 2002 - 20:36:18 CDT

On Wed, Aug 07, 2002, Giulio Rastelli wrote:
>
> I am running an MD simulation of a protein using the GB-SA model (amber7 on
> a IBM-SP3 computer).
> Unexpectedly, the simulation works fine when I let everything free to move.
> Instead, when I define a subset
> of residues allowed to move (ibelly=1) the residues of the belly shift more
> the 60Angs apart from the rest
> of the protein, with consequent blow up of the energy! Is it correct to use
> belly with GB-SA?

The use of belly with GB is very limited. Unfortunately, the test that is
made to see that things are OK (basically, that the moving atoms are at
the beginning of the system) is mistakenly being skipped (due to a conflict
with the decomposition code...)

Change line 1238 in mdread.f from

      if( belly .and. igb.gt.0 ) then
to
      if( ibelly.gt.0 .and. igb.gt.0 ) then

This will give a correct error message if you try to do what you have done.

Using a belly approximation with GB is a tricky business (involving refitting
parameters), and this has not yet been done with Amber. For some persepctive,
see:

%A O. Guvench
%A J. Weiser
%A P. Shenkin
%A I. Kolossvary
%A W.C. Still
%T Application of the Frozen Atom Approximation to the GB/SA Continuum Model
for Solvation Free Energy
%J J. Computat. Chem.
%V 23
%P 214-221
%D 2002

I'll post a fix to the Amber web site....thanks for reporting this problem.
Sorry that I can't give a better response, but at least the program should
terminate now with an error message.

..regards...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
==================================================================