[gmx-users] Free Energy Change of Growing a Molecule

Justin Lemkul jalemkul at vt.edu
Fri Oct 10 20:16:37 CEST 2014



On 10/10/14 10:45 AM, Barnett, James W wrote:
> Good morning,
>
>
> I'm looking at doing a free energy change with thermodynamic integration with
> GROMACS. Specifically I want to "grow" a molecule and find out the difference
> in the free energy. For example, I would like to see the free energy change
> between a methane and ethane.
>
>
> I've looked at Dr. Lemkul's tutorial about free energy
> changes<http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin/gmx-tutorials/free_energy/index.html>
> and he refers to the manual regarding changing a molecule with a dual-toplogy
> approach. I've reviewed how to use two topologies in the manual in section
> 5.7.4. In the manual's example the number of atoms in the molecule remained
> the same from topology A to B.
>

Well, that's only true because it's a united-atom force field, so it's rather 
convenient.

>
> My questions are:
>
> 1.  Is it possible to calculate the free energy change as I am describing,
> where the number of atoms in the molecule changes from topology A to B? 2.
> If it is possible, how is this implemented in the dual topology approach?
>
> I did find this
> thread<https://mailman-1.sys.kth.se/pipermail/gromacs.org_gmx-users/2005-May/015019.html>
> in the mailing list archive where someone basically answered "no", but the
> post is almost 10 years old and a few things have changed with GROMACS since
> then.
>

The reason given there has nothing to do with whether or not Gromacs can carry 
out the calculation (it can), rather there may be some fundamental problems with 
the approach.  I can't speak to how severe those might be, but people certainly 
carry out simulations of "mutating" one residue into another quite often.  It's 
just a matter of transforming dummy atoms into real ones, or vice-versa.  There 
is a wealth of information on this topic at www.alchemistry.org, as well as 
practical examples.  I would suggest you look there.  Relative binding affinity 
(mutating phenol to toluene, note different numbers of atoms!) is one topic that 
is covered.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================


More information about the gromacs.org_gmx-users mailing list