Re: [AMBER] MMGBSA Decomp on Heme Systems

From: Jason Swails <>
Date: Tue, 24 Nov 2015 12:24:38 -0500

On Tue, Nov 24, 2015 at 11:46 AM, Vivek Bharadwaj <
> wrote:

> Dear Amber users,
> I am trying to perform binding energy decomposition energies for a
> substrate bound protein system that has a Heme bound. I am using Amber14
> with Ambertools15.
> I understand that Sander requires the radius for Fe atom which requires
> modification of the mdread2.F90 file and recompiling of sander.
> I followed the instructions based on this previous query (
> and modified the mdread2.F90
> file and also recompiled sander. I also re-generated the complex, ligand
> and receptor topologies after recompiling amber.

​Regenerating the topology files is unnecessary.

> However, I still get the following error while performing decomp
> calculations. I am able to run normal mmgb/pbsa calculations without decomp
> ​​
> though.
​Are you sure you modified the correct part of mdread2.F90? There are two
locations where these radii need to be set. The first is when gbsa=1
(which I believe occurs first in the source code file) and the second is
where gbsa=2. Decomposition uses gbsa=2. So I suspect you needed​

​to make this change in a second place in order for it to work for your


Jason M. Swails
Rutgers University
Postdoctoral Researcher
AMBER mailing list
Received on Tue Nov 24 2015 - 09:30:03 PST
Custom Search