AMBER Archive (2009)

Subject: Re: [AMBER] question about pair wise per-residue energy decomposition; idecomp=3

From: case (
Date: Wed Mar 25 2009 - 13:58:02 CDT

On Wed, Mar 25, 2009, wrote:
> I'm interested in the energy decomposition with the pairwise per-residue
> manner. From the AMBER mannual, I found the SANDER in AMBER can do this
> job if we set idecomp=3 or 4. If we set GBSA=2, the Esurf (nonpolar
> solvation energy) is also decomposited into pairwise form.
> My question is how the Esurf can be decomposed into pairwise per-residue
> form since this term is not a real pairwise energy.

Each surface point is assigned to a particular atom, so the total surface area
(and hence Esurf) can be treated as the sum of individual atomic terms. This
is "pairwise" in the sense that the diagonal terms are each atom's
contributions, and all the off-diagonal (interaction) terms are zero.

...hope this helps...dac

AMBER mailing list