Re: Free energy calculation yields strange results

From: Jérôme Hénin (jhenin_at_ifr88.cnrs-mrs.fr)
Date: Mon May 10 2010 - 08:04:43 CDT

Hi,

The most obvious problem here is the energy minimization before each
FEP stage. It is not recommended, as 500 MD steps will not be
sufficient to thermalize the system back to room temperature. There
should be no need for minimization there.

In addition, there could possibly be a mistake in your setup, or a
(large) convergence/sampling problem. Interactions between ions are
notoriously difficult to sample:
http://www.ncbi.nlm.nih.gov/pubmed/15584080

That is all I can say based on the information you provided. One test
you can run is the same transformation in vacuum (remove the PBCs and
forget about neutralizing the charge).

Cheers,
Jerome

On 9 May 2010 14:53, Jun Zhang <coolrainbow_at_yahoo.cn> wrote:
> Hi Everyone!
>
> I am perfoming alchemical free energy calculation using NAMD 2.7b2. The transformation of the species is from chirality R to S
>
> A(R) --- A(S)
>
> in TIP3P water molecules. The box is about 50A. The species has a COO and I have use some ions (Na+ and Cl-) to neutralize it. I think the free energy should be ZERO since water is achiral. However, my result is 2328.95
> kcal/mol, which made me surprised. My alchemical parameters set is:
>
> alch on
> alchType FEP
> alchFile R2S_wb.fep
> alchCol B
> alchOutFile R2S_forward.fepout
> alchOutFreq 10
>
> alchVdwLambdaEnd 1.0
> alchElecLambdaStart 0.5
> alchVdWShiftCoeff 6.0
> alchDecouple yes
>
> alchEquilSteps 500
> set numSteps 5000
>
> set Lambda0 0.0
> set dLambda 0.05
>
> while {$Lambda0 <= 1.0} {
>
> alchLambda $Lambda0
> set Lambda0 [expr \$Lambda0 + \$dLambda]
> alchLambda2 $Lambda0
> minimize 500
> run $numSteps
>
> }
>
> Can anyone give some suggestions? Thank you in advance.
>
>
> Jun Zhang
> Nankai University
> coolrainbow_at_yahoo.cn
>
>
>
>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:55:46 CST