VMD-L Mailing List

From: Josh Vermaas (vermaasj_at_msu.edu)
Date: Wed Mar 06 2024 - 15:15:26 CST

Hi Jacob,

Why not just align the protein before doing the analysis? Then you could
use geometric criteria. The psuedocode would look something like this:

#Load trajectories

#Align protein to frame 0 (or whatever your favorite frame is)

#Make a selection
set sel [atomselect top "seltextgoeshere"]
#Update this selection every frame, count the number of elements.
for { set f 0 } { $f < [molinfo top get numframes] } { incr f } {
$sel frame $f
$sel update
puts "Frame $f: [$sel num]"
}

-Josh

On 3/6/24 2:43 PM, Jacob Layton wrote:
> Hello,
>
> The basic idea is to count the number of water molecules inside the
> water channel of my protein for each frame of a trajectory. The issue
> is that the protein is not fixed in space, and simply defining the
> area using selections such as "x > -10 and x < 10 and y > 20 and y <
> -20 and z > 25 and z < -25" (used in one of my previous counts)
> doesn't account for the protein's movement and will eventually include
> waters not present in the channel.
>
> I'm not sure if it would be possible using tcl scripting to assign
> certain residues as "boundaries" for the water channel region. Or
> perhaps adding a dummy atom at the center of mass of the region that
> would move alongside the trajectory. I did notice the gromacs water
> transition count tool from a previous thread here
> (https://www.ks.uiuc.edu/Research/vmd/mailing_list/vmd-l/25483.html
> <https://urldefense.com/v3/__https://www.ks.uiuc.edu/Research/vmd/mailing_list/vmd-l/25483.html__;!!HXCxUKc!1G5wqLXQuBmxqqRhJEXh7VZEuCpHEJsk2VXrDnOQRDJ9v6WuUf69Q8s8fC4VsLNUuXakFq_KeE4v9-2C5g$>)
> and have been contemplating whether it would be worthwhile to convert
> my namd files. Any insight would be appreciated. 

-- 
Josh Vermaas
vermaasj_at_msu.edu
Assistant Professor, Plant Research Laboratory and Biochemistry and Molecular Biology
Michigan State University
vermaaslab.github.io