Umbrella Sampling Replica Exchange Simulations

From: TREIKALIS Antons (
Date: Sat Jul 13 2013 - 11:24:12 CDT

('binary' encoding is not supported, stored as-is)

Dear NAMD developers,

My question is regarding the stretch alanin example from the NAMD_2.9_Linux-x86/lib/replica/umbrella directory. I have managed to successfully run the simulation, but I am required to understand what actually is happening during the simulation. Can you please answer these questions:

 - Data from which files is used for exchanges between replicas? (for this particular example or in more general terms)
 - What is actually exchanged between pairs of replicas?
 - Is all data/parameters required for exchanges between replicas stored in intermediate files, or there are some internal NAMD data structures used for exchanges? In other words, is it enough to obtain necessary data from the output files of another replica, insert this data in appropriate files/places and to restart the simulation?
 - What are the rules for determining which replicas will exchange the data? Algorithm?
 - For this particular example, can you please provide more information on data flow?

I apologise if this sounds a bit naive, but I have zero experience in MD simulations. My intention is to use NAMD as an MD engine for the framework I am currently working on (MSc project).

Thank you very much for your help.

Kind regards,

The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:23:27 CST