VMD-L Mailing List

From: Giacomo Fiorin (giacomo.fiorin_at_gmail.com)
Date: Thu Jan 24 2019 - 21:30:23 CST

Dear all, there is a bug in the volmap utility, which manifests itself when
"density" or "inter" are computed over a selection that changes between
trajectory frames (e.g. the water molecules "within" some distance from a
protein residue). This is caused a faulty allocation of the array of
weights, which was set as long as the number of atoms in the selection *for
the frame that was active when volmap was invoked*. Even with the default
weights (1 for each atom), atoms that are more numerous than in the initial
frame may read outside this array, and will have an undefined weight
(sometimes zero, sometimes a completely random number aka "garbage").

A fix has just been merged by John Stone into the main VMD distribution,
and will be available in the next alpha or beta release. Until then, some
of you may be interested in recompiling VMD locally from source.

For VMD 1.9.3, a patch is available in the v1.9.3 branch of this repository:
https://github.com/giacomofiorin/vmd-patches/branches
In the master branch, which follows VMD CVS, the same patch is empty
because the fix has already been included.

A test script is also available in the same repository to diagnose the bug
(or its absence). Feel free to tweak it to your own use case to check if
you need to update.

Lastly, please note that newer and better-performing commands are coming
into VMD to achieve similar goals to volmap. John and the rest of the UIUC
team will provide more information about them when they become available.

Giacomo 

-- 
Giacomo Fiorin
Associate Professor of Research, Temple University, Philadelphia, PA
Contractor, National Institutes of Health, Bethesda, MD
http://goo.gl/Q3TBQU
https://github.com/giacomofiorin