AMBER Archive (2004)

Subject: Re: AMBER: replica exchange with thermodynamic integration

From: David A. Case (case_at_scripps.edu)
Date: Fri Sep 24 2004 - 10:44:01 CDT


On Thu, Sep 23, 2004, david.evans_at_ulsop.ac.uk wrote:
>
> I'm curious about the option to use MMTSB in combination with lambda
> replica-exchange (pg 150 Amber8 manual). I have a copy of the MMTSB
> toolkit, but cannot find documented anywhere exactly what this option
> does e.g.what is the acceptance probability used for exchange moves
> between replicas with different lambda values. (My vague plan is to
> run TI calculations at different lambda values in parallel, and use a
> free energy perturbation-style exchange move to hopefully enhance
> sampling.)
>
> Could anyone with experience of using this tool explain this in more
> detail, or give a reference?
>

This is an idea that we also have had, and that influenced how the MMTSB
hooks were placed into Amber. My memory is that there was some very
preliminary testing of this, but it never got very far -- which is why the
documentation probably won't help very much.

Scott Brozell was the person who worked on this -- maybe he will have some
comments or recollections. I'll cc this e-mail to him. We will certainly try
to help out if we can.

...dac

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu