AW: Temperature assignment REMD

From: Norman Geist (norman.geist_at_uni-greifswald.de)
Date: Mon Jan 05 2015 - 01:29:55 CST

If you use explicit solvent, you might btw. be interested in the TIGER2 REMD algorithm which overcomes the problem of many needed replicas for suitable exchange probability and the problem of raised exchange probability on higher temperature replicas due increased pressure and small potential energy difference.

 

I have made a implementation suitable for NAMD 2.10 based on the MPI calls. If you want, or anyone else I could post it here.

 

Norman Geist.

 

Von: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] Im Auftrag von divya nayar
Gesendet: Samstag, 3. Januar 2015 15:36
An: namd-l_at_ks.uiuc.edu
Betreff: namd-l: Temperature assignment REMD

 

Hi,

 

I am doing REMD simulation of small peptides in explicit solvent using NAMD 2.9 version. I want to know more about the way NAMD generates temperatures for replicas. It uses an exponential formula, but I want to know the reason as to why NAMD uses it. Is there any reference from literature for the same because I have come across other methods of generating temperatures based on potential energy distributions of the system which have been described in literature .

 

What is the advantage of using exponential distribution of temperatures for replicas of system?

 

Best,

Divya

This archive was generated by hypermail 2.1.6 : Tue Dec 27 2016 - 23:20:46 CST