Re: [AMBER] About turning off energy terms between selected atoms

From: Jason Swails <jason.swails.gmail.com>
Date: Thu, 21 Nov 2013 07:44:31 -0500

On Thu, 2013-11-21 at 07:33 -0500, David A Case wrote:
> On Tue, Nov 19, 2013, Mo Chen wrote:
> >
> > I would like to run an MD simulation, during which the nonbond interaction
> > energy between certain atoms are turned off. I would assume perhaps the
> > specified atoms can be masked some way, but I have not been able to find
> > much detailed information about this. May I ask how this can be achieved?
> > Is there a similar function as the BLOCK function in CHARMM?
>
> We don't offer explicit support for such things. Hacking the non-bonded
> exlusion list ought to do the trick, but you'd have to expect to take some
> time in (first) understanding how to do that; and (second) doing some careful
> checks to make sure you are getting what you want.

This will work in GB and vacuum, but it is harder with PME. You can use
ParmEd to add arbitrary exclusions with the "addExclusions" command. It
goes something like this:

parmed.py -p <prmtop>

addExclusions <mask1> <mask2>
outparm exclusion.prmtop

This command adds every atom in <mask2> to the exclusion list of
<mask1>. As I mentioned above, though, only the GB code actually uses
the exclusion list in the topology file. The PME code regenerates the
exclusion list from the bond and angle network.

You could hack the exclusion list by adding bonds between all of the
atoms you want excluded, with a bond force constant of 0. The "setBond"
command in ParmEd will do this for you (but it will be quite tedious
since you need to add the bonds individually).

HTH,
Jason

-- 
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Nov 21 2013 - 05:00:03 PST
Custom Search