[gmx-users] User Defined Potentials

[Berk Hess]

>From: David van der Spoel <spoel at xray.bmc.uu.se>
>Reply-To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>To: Discussion list for GROMACS users <gmx-users at gromacs.org>
>Subject: Re: [gmx-users] User Defined Potentials
>Date: Tue, 13 Sep 2005 08:31:20 +0200
>
>On Mon, 2005-09-12 at 15:51 -0500, hassanal at purdue.edu wrote:
> > Gromacs Users,
> >
> > I would like to insert an "additional" potential of the form 10-12 
>between 2
> > specific atoms in my system to model the underestimated pi-cation 
>interaction
> > purely given by a LJ 6-12 interaction. I understand that user defined 
>potentials
> > are put in table format but I can't see any description on how to 
>specify the
> > atom  pair that it will be applied to. It looks as though the user 
>defined
> > potential will automatically be applied to all LJ pairs in the topology 
>file. Is
> > this a correct observation or not?
> >
> > Would it be possible to add a potential of the form 10-12 (without going 
>into
> > the guts of gromacs code) and to add it to only 1 pair of atoms in my 
>system?
> >
>How about using a Morse bond instead?
>
>What you ask is not trivial...

The new Gromacs release supports different user potentials for different
pairs of energy groups. So there it would be possible by putting the two
atoms into a seperate energy group.

Berk.