AMBER Archive (2009)

Subject: Re: [AMBER] molecular crystal simulation

From: David A. Case (
Date: Wed May 13 2009 - 17:28:37 CDT

On Tue, May 12, 2009, Simona Irrera wrote:

> I performed a 1ns simulation (NPT) of a molecular crystal (I modified
> the initial coordinate in the inpcrd with the angles alfa/=beta/=gamma
> etc). I want to explore all the possible configurations also existence
> of different polymorphs.
> In the geometry coordinates in the file.mdcrd I can't see the angles,
> does it mean that during the simulation the box is fixed in angles and
> only the a, b and c dimensions change?

I think this is correct, and that Amber does not have the ability to modify
the shape of the unit cell in constant pressure simulations. Others can
correct me if I am wrong.

Also bear in mind that you need a big unit cell, such that the nonbonded
cutoff used is less than half the shortest box dimension. You might need to
put several real unit cells together to satisfy such a requirement.


AMBER mailing list