From: Chris Chipot (Christophe.Chipot_at_edam.uhp-nancy.fr)
Date: Wed May 11 2005 - 17:34:47 CDT
There is, indeed, a singularity when lambda tends towards
0 or 1 with growing particles - known as "end-point"
catastrophe. One way to circumvent this difficulty is to
use a soft-core potential in the spirit of Beutler et al.
Chem. Phys. Lett. 1994, 222, 529-539. Unfortunately, we
have not implemented it yet in NAMD. A work-around that you
might want to try is to use windows of uneven width, getting
infinitesimal as you approach the end points of the reaction
path. In addition, you may want to equilibrate and start at
lambda = 0.5, and move in both directions, towards 0 and 1,
to avoid poor vdW contacts due to growing atoms (note that,
strictly speaking, in the dual-topology paradigm, we scale
the interaction potential energy with lambda rather than the
non-bonded parameters of the appearing/disappearing side chains).
Wei Chen wrote:
>Now I am using free energy perturbation to calculate free energy difference.
>The system is one ligand + one receptor. I want to mutate one
>residue in binding pocket from Glycine to Aspartate.
>Basically, I grow aspartate and shrink glycine at the same time.
>The problem is that aspartate could overlap with other residues due
>to its bigger size than glycine. So the VDW energy becomes very large,
>even infinite. Anybody has ideas to deal with this situation?
Chris Chipot, Ph.D.
Equipe de dynamique des assemblages membranaires
Unité mixte de recherche CNRS/UHP No 7565
Université Henri Poincaré - Nancy 1 Phone: (33) 3-83-68-40-97
B.P. 239 Fax: (33) 3-83-68-43-87
54506 Vandoeuvre-lès-Nancy Cedex
To sin by silence when we should protest makes cowards out of men
Ella Wheeler Wilcox
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:40:45 CST