[gmx-users] topology file and LJ parameters

[David Mobley]

David,

I wrote back with a question and didn't get an answer, so let me rephrase.

If YOU were going to try and turn off only intermolecular vdW interactions
(maintaining intramolecular), how exactly would you set up your topology
file to do this?

I tried to do it as I describe below. Do you see anything obviously wrong
with my setup (which doesn't seem to be working)?
1) Define new atom types with the same masses, but with no vdW parameters
2) Change the atoms listing to add perturbations to these atom types in the
typeB column.
3) Add pair interactions between each of the atoms (with non-excluded
interactions; here nrexcl=3) in my molecule explicitly in the pairlist,
including explicit statement of the interaction parameters.

For some reason if I do the above I am still getting a nonzero dv/dlambda in
vacuum, which means I'm somehow changing the intramolecular interactions
(right?), even though I don't think I should be.

I can provide a sample topology file if that would help.

Thanks,
David


On 11/26/05, David van der Spoel <spoel at xray.bmc.uu.se> wrote:
>
> David Mobley wrote:
> > Dear all,
> >
> > I'm doing a free energy calculation which involves turning off a small
> > molecule's interactions with its environment. Ordinarily, for the vdW
> > part of this, I simply turn off the vdW interactions of the small
> > molecule. However, it would be better to turn off the vdW interactions
> > with the environment, ONLY. Someone had mentioned to me it should be
> > possible to do this using careful manipulation of the topology file in
> > GROMACS, and I've been trying to figure out how to do it, but have
> > been unsuccessful.
> You mean you want to maintain the intramolecular VdW interactions?
> In that case you could specify all the interactions as pairs, with
> parameters different from the normal LJ.
> It's only 200...
>
> >
> > Can anyone recommend a way to do this? Here's a couple of ideas I'd
> > come up with that I haven't been able to get to work so far (by
> > "work", I mean that I've been testing them in a case where I try
> > turning off the small  molecule's interactions in vacuum, where
> > dv/dlambda for turning off vdW interactions should be zero if I'm
> > leaving the intramolecular vdW interactions on; when they don't work,
> > the result for this is NOT zero):
> >
> > (a) Set up normal and perturbed atom types as usual (where the
> > perturbed atom types have vdW turned off) but then specify nonbonded
> > interactions for the atom types that I want to have continue to
> > interact with one another. For some reason if I do this and have a
> > pairlist for 1-4 interactions, I get big contributions to dv/dlambda
> > from the 1-4 terms even if I don't actually perturb any of the vdW
> > parameters (i.e. if I only change the atom types to equivalent ones).
> > (b) Specify in a pairlist all of the pairs I want to have continue
> > interacting, with explicit statement of parameters.
> >
> > Should it be possible to make this work? If so, can anyone give me
> > some guidance how it ought to work?
> >
> > I would also note that my small molecules are quite small (i.e. 20
> > atoms) so I'm not opposed to solutions which require explicit
> > enumeration of all of the interacting pairs within these small
> > molecules.
> >
> > Thanks,
> > David
> > _______________________________________________
> > gmx-users mailing list
> > gmx-users at gromacs.org
> > http://www.gromacs.org/mailman/listinfo/gmx-users
> > Please don't post (un)subscribe requests to the list. Use the
> > www interface or send it to gmx-users-request at gromacs.org.
>
>
> --
> David.
> ________________________________________________________________________
> David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
> Dept. of Cell and Molecular Biology, Uppsala University.
> Husargatan 3, Box 596,          75124 Uppsala, Sweden
> phone:  46 18 471 4205          fax: 46 18 511 755
> spoel at xray.bmc.uu.se    spoel at gromacs.org   http://xray.bmc.uu.se/~spoel
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> _______________________________________________
> gmx-users mailing list
> gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-request at gromacs.org.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20051128/43bec7cd/attachment.html>