AMBER Archive (2006)

Subject: Re: AMBER: How the ensemble average calculated in Gibbs, AMBER7?

From: David A. Case (case_at_scripps.edu)
Date: Sun Nov 05 2006 - 21:26:02 CST

On Wed, Nov 01, 2006, Yongmei Pan wrote:
>
> Just one question about how the ensemble average is calculated in the
> following equation:
>
> G(i+1)-G(i)=-RTln<exp-[(V(i+1)-Vi)/RT]>i
>
> for the samplings in states i and i+1, how the potential energie V(i+1)
> subtracted by Vi? Because we can not say, for example, the No.100 sampling
> from state i+1 corresponds to the No. 100 sampling from state i, how can we
> calculate the ensemble average of V(i+1)-Vi?

Within the amber suite of programs, you cannot calculate the required averages,
for the reasons you cite. Amber currently supports thermodynamic integration,
but not free energy perturbation (FEP). The equation above is for FEP. A code
change (not all that hard, by the way) would be necessary to support FEP.

...hope this helps....dac

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