RE: PMF decompostion

From: Radak, Brian K (
Date: Mon Sep 12 2016 - 12:31:48 CDT

I have not worked with the pair interaction utility, but if the selection is done in the B column before the calculation, I doubt that a dynamic selection is possible, since those assignments are only done when the "structure" command is called.

You might be able to hack this by swapping the coordinates of certain water molecules via Tcl in between energy evaluations, but that assumes: 1) you can determine which molecules you want on-the-fly (maybe via "measure"?) and 2) the number of atoms in the selection is fixed. The validity of number 2 probably depends on how you define the first solvation shell.


Brian Radak
Postdoctoral Appointee
Leadership Computing Facility
Argonne National Laboratory

9700 South Cass Avenue, Bldg. 240
Argonne, IL 60439-4854
(630) 252-8643
From: [] on behalf of Azadeh Alavi []
Sent: Monday, September 12, 2016 6:49 AM
To: NAMD list
Subject: namd-l: PMF decompostion

Dear Namd users

I have a membrane-protein system using charmm36 force field and have calculated the PMF profile for the permeation of an ion across the channel using the ABF method. I wished to know how to decompose the PMF profile into individual contributions from the ionís interactions with its first solvation shell water (defined as 3.3 Aį from sodium for instance), using pair interaction utility in NAMD. As far as I know, in the this utility, a file should be prepared in which selected atoms are to be distinguished by 1 and 2 in the B column. However, in order to consider the first solvation shell water, the file needs to be updated while the .dcd file is being read and I do not know how it it is possible.

Your help would be greatly appreciated.
Thank you.

Best regards,

This archive was generated by hypermail 2.1.6 : Tue Dec 27 2016 - 23:22:27 CST