[gmx-developers] Re: issue on benchmarking FEP calculations in Gromacs and NAMD

[Igor Leontyev]

Hello,
Recently I did FEP calculations for Na+ by Gromacs and Amber packages and
found excellent agreement. Please, make sure that there is no miscounting of
the Ewald correction in your data because the results suspiciously differ by
the exact value of the correction (~11.5 Kcal/mol). As I know Gromacs and
other common codes do not include the correction in their output. Aslo try
to recalculate free energy by simple slow growth method: init_lambda=0;
delta_lambda=1/nsteps.

Igor Leontyev

> Hi,
>
> What is the correction you applied?
This is Ewald correction suggested by G. Hummer and co-authors J. Phys. 
Chem. 1996, 100, 1206 to extract from output the interaction  energy of the 
solute with own periodic images. This is necessary to compare the simulation 
with periodic boundary conditions (in which the solute interacts with 
itself) with experimental data corresponding to infintly deluted solutions.

> I think Gromacs has all charge corrections in, except for the set of
> corrections
> required for absolute solvation free energies as described by Philippe
> Hünenberger.

The Ewald correction is not applied in Gromacs. Otherwise, the charging 
energy of ion in empty box (no solvent) would be zero which is not true.

> Cheers,
>
> Berk