[gmx-developers] pairs types for decoupling

[David Mobley]

All,

I'm using the CVS version of 3.3.1 from last week, and am trying to
get the new pairs types Berk Hess has implemented to work for doing
decoupling. There's an old e-mail here on the subject, which is all
I've got to go by:
http://www.gromacs.org/pipermail/gmx-developers/2006-March/001530.html

Berk wrote this:

>The idea is that for the solute to be decoupled one would replace
>the in its [ pairs ] section type 1 by type 2
>and exclude all intramolecular non-bonded interactions and replace
>add them again explicitly in a [ pairs ] section as type 3.
>...
>BTW I tested it and it works.

Berk, any chance you can provide your test files so I can see how
exactly you're doing this?

I've been trying to do this for toluene, using Michael Shirts'
parameters which were published in supporting information; I've
previously reproduced his values without using decoupling so I know
they're correct. So far, I'm getting stuff that's ridiculously wrong
when I attempt decoupling, so I must be doing something wrong. Here's
what I'm doing:
1) I added bonds type 5 between all of the atoms that don't have
bonded interactions to ensure that the non-bonded interactions are
excluded.
2) I changed all of the existing pairs to pairs type 2 (without
providing parameters, which I assume will mean they're still generated
according to teh combination rules for the A state)
3) Put all intramolecular nonbonded interactions into the pairs
section using pairs type 3 (again without providing parameters, since
they should be generated using combination rules for the A state).

What am I doing wrong? It would probably be very useful if one of us
put, either online or in the manual, a simple example topology that
demonstrates how to do this.

Thanks,
David