[AMBER] Solution to the replica exchange SGLD restart bug

From: Charo del Genio <the.paraw.gmail.com>
Date: Mon, 1 Feb 2021 21:40:41 +0000

Dear all,
        after some more work, I finally produced some patches that work for the problem of the restart of RXSGLD runs. I verified that they are correct by checking trajectories literally step by step and
outputting a certain amount of debug information.

Please find attached the patches (for sander, remd and getcor). The idea is that now getcor detects whether there is a stage ID in the restart files and, if so, propagates it to remd1d_setup, which
then takes care for changing tempsg and sgft to the correct values for that stage.

Right now, I am virtually sure that it works correctly. However, once more, I would greatly appreciate comments from the developers, because it is always possible that I made some subtle mistake.


Cheers,

Charo




-- 
Dr. Charo I. del Genio
Senior Lecturer in Statistical Physics
Applied Mathematics Research Centre (AMRC)
Design Hub
Coventry University Technology Park
Coventry CV1 5FB
UK
https://charodelgenio.weebly.com





_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber

Received on Mon Feb 01 2021 - 14:00:02 PST
Custom Search