AMBER Archive (2009)

Subject: Re: [AMBER] please help me out

From: David A. Case (
Date: Wed Apr 08 2009 - 13:10:04 CDT

On Wed, Apr 08, 2009, Hugh Heldenbrand wrote:
> I have recently been doing many hundreds of minimizations of a small
> molecule systems, and I see this error for probably one out of every
> hundred minimizations that I do. I won't comment on the stability of a
> protein in the gas phase, but my solution to this particular error is to
> restart the minimization with a number steps smaller than the number
> that it took to fail. This will successfully produce a restart file
> that I can use to keep minimizing for however many steps that I want to
> (presumably the second minimization does not exceed its new virtual box).

These problems can be avoided by using igb=6, although that might be a
slower option (depends on the problem). It's certainly worth a try for
system that fail.

The entire idea of having a virtual box around vacuum systems has too
many ways to fail....


AMBER mailing list