Re: problem w/ free energy perturbation

From: Jerome Henin (heninj_at_gmail.com)
Date: Wed Nov 29 2006 - 20:04:53 CST

Hi Omar,
Actually, it is very likely that these clashes occur between 'mutant' atoms
and the rest of the system, for instance if lambda is very close to zero. Try
outputting a DCD trajectory file with a very high frequency and visualizing
the last recorded frame. Is there no overlap happening except between
mutating atoms?
If no, there might be a slight mistake in the fepFile resulting in some
duplicated atoms not being declared as 'wild type' or 'mutant'. Try loading
the fepFile as PDB in VMD, and setting the coloring method to Beta to display
these groups.
Jerome

On Wednesday 29 November 2006 19:51, OMAR NABEEL ALY DEMERDASH wrote:
> Hello,
> I am a novice to the free energy perturbation method. I have a system in
> which I gradually turn off the charges of a particular residue. I get the
> following error message:
>
> ERROR: Margin is too small for 252 atoms during timestep 1.
> ERROR: Incorrect nonbonded forces and energies may be calculated!
>
> It seems like this may be due to the presence of nonbonded interactions
> between the 'wild type' atoms and the 'mutant' atoms. (The vdW energy was
> unrealistically large:99999999.9999.) However, this should not be
> happening, because I ran the alchemify program, which is supposed to
> specify the exclusions.
> Thanks in advance!
>
> Omar Demerdash

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:44:13 CST