[gmx-developers] energy distribution

[Robert Vacha]

Sorry, I'm not sure I understand what criterion you mean. I thought,
that I would do it as interaction energy is calculated and written in
energy file now (If energy groups are set, gromacs write in energy file
LJ and coulomb between the selected groups, and I would like to add the
histogram of the sum of LJ and coulomb). So the criterion would be just
presence of molecules in selected group, so for all of them the standard
Lennard-Jones interaction limited with cut-off plus coulomb interaction
with Ewald would be calculated. Of course, there could be molecules
which would have interaction energy equal 0, but that shoudn't be a
problem, right? Do you see any problem with calculating what I described
above?

thanks,
Robert

On Wed, 2008-04-09 at 10:09 +0200, Anton Feenstra wrote:
> Robert Vacha wrote:
> > Yes indeed, I'd like to calculate histogram in energy of molecular
> > interactions. It's possible do it now by defining the solute molecule
> > and then separately all solvent molecules as energy groups, writing
> > interaction energies for all of them them and then making histograms,
> > but this is very slow and unpractical. That's why I'd like to implement
> > it either to the main run - (one would set in mdp input that he wants
> > histogram in interaction energies of some groups plus the size of
> > histogram grid) or in the separate script (this made sense only if it's
> > possible easily call some gromacs functions that would calculate the
> > interaction energies).
> 
> So then what criterion would you use to decide when there is an 
> interaction and when not? This would be critical, and would influence 
> how interesting this feature might be to other users.
> 
> Groetjes,
> 
> Anton