Re: alchemical fep amber topology

From: Floris Buelens (floris_buelens_at_yahoo.com)
Date: Sat Mar 14 2009 - 06:16:37 CDT

Sorry, I answered George's email before reading this reply. I have been confused in the past as to why Alchemify bothers to add non-bonded exclusions, since according to my reading of the code (going back to the original FEP in version 2.4) and experience, non-bonded interactions between the two lambda-coupled groups have always been excluded automatically at the source code level. Am I missing something fundamental here?? Do the non-bonded exclusions that alchemify adds impact on anything other than the pairwise force / energy calculation?
Back to the question, I use Amber format paramater/topology files with FEP and they're definitely fine with any groups that aren't bonded to the rest of the system, however if you wanted to do something with bonded groups some custom post-processor along the lines of alchemify would be required to remove angles and dihedrals that link that lambda=0 and lambda=1 end points.
Best,

Floris

----- Original Message ----
From: Jerome Henin <heninj_at_gmail.com>
To: George Madalin Giambasu <giambasu_at_gmail.com>
Cc: namd-l_at_ks.uiuc.edu
Sent: Friday, 13 March, 2009 20:24:43
Subject: Re: namd-l: alchemical fep amber topology

Hi George,

With NAMD 2.6, you need Alchemify and a PSF file. The program CHARMM
is not needed, but you have to use its file format (typically created
using psfgen). I don't know if it is possible to define custom
nonbonded exclusions in an AMBER parmtop file in a way that NAMD 2.6
would understand. I have never tried.

If, however, you obtain and compile the current development version of
NAMD (or any future release), you won't need to worry about nonbonded
exclusions and all that sort of things. That is now taken care of
internally.

Best,
Jerome

On Fri, Mar 13, 2009 at 2:55 PM, George Madalin Giambasu
<giambasu_at_gmail.com> wrote:
> Hi,
>
> I was wondering if there is a way to run alchemical fep simulations
> using Amber topologies (*.parm7 files) using the cvs version of namd?
>
> I found a post by Jerome Henin
> (http://www.ks.uiuc.edu/Research/namd/mailing_list/namd-l/1348.html) on
> the list in which it was stated that the fep setup "does not need CHARMM
> anymore", however the Alchemical FEP tutorial states that we have to use
> psf files in conjunction the alchemify program.
>
> Thanks in advance for any suggestions or updates!
>
>
> George
>
>

      

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 05:21:49 CST