Re: Simultaneous FEP simulation of two distinct residues

From: Jérôme Hénin (
Date: Wed Dec 15 2010 - 04:04:47 CST

Hi Christian,

On 14 December 2010 20:44, Christian Jorgensen
<> wrote:
> Hi all,
> In a normal free energy perturbation simulation, one selection/group is simulated for a creation/annihilation event that yields a free energy change deltaG.
> If two residues are studied, currently two options exist:
> I) Run two simulations, one for each residue, and obtain two free energy values deltaG(1) and deltaG(2)
> II) Run one simulation and obtain an average over the two groups deltaG, where deltaG= (1/2)[deltaG(1) + deltaG(2)]. The disadvantage is that the free energy value is only an average value, not entirely representative of the two distinct residues.
> I have a dipeptide Glu-GluP that I wish to study in a single simulation. I wish to create a proton in Glu and annihilate the proton in GluP.
> I want to study this in a single simulation, but I do not want the average value over the two groups. I want to obtain the distinct values pertaining to the two groups studied deltaG(1) and deltaG(2)
> My question: Is there a way of configuring the NAMD FEP configuration file and/or hybrid PDB file to study two distinct creation/annihilation processes during one simulation and obtain the free energy values for the two processes?



PS: As usual, the FEP results only tell you something about
interactions with the environment (solvation free energy), and nothing
about covalent interactions. If your goal is a relative pKa
calculation, you will have to run reference calculations with e.g.
individual aminoacids in water, and introduce the experimental pKas
for those.

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:54:51 CST