From: Axel Kohlmeyer (
Date: Sun Mar 14 2010 - 10:10:02 CDT

On Sun, 2010-03-14 at 10:51 +0100, Francesco Pietra wrote:


as i already wrote in response to your previous query, your
"handwaving" description is not very helpful to track down
your problem, and the fact that you can reproduce it, doesn't
make it a more convincing issue. i can only speculate that
your problem is caused by a selection that includes unwanted
atoms. all the script does is to draw lines between the center
of mass coordinates for the selection. thus there is little
possibility of a bug in the script code and high probability
of a PEBCAC situation.

in case you don't believe this, the only way to prove it to
you, would be if you can provide a sufficiently large chunk
of your trajectory data and the exact script code to reproduce
what you are seeing. due to the size of files, the recommended
way to make those available is to upload them to the VMD biofs
via an account on biocore.


> I have carried out many runs for the exit of the ligand from the
> protein. I can confirm that, with the commands described previously
> (please see below previous mail), the trace does not start from where
> the ligand starts leaving. It starts later on and not exactly on the
> actual path, as monitored from VMD Main, once, step 5, speed slow. In
> the most complex path for ligand exit, the trace (red on "scale")
> starts on the rear of a helix with respect to the initial position of
> the ligand. I.e., the initial position of the ligand can't be seen
> from where the trajectory_path trace begins. Also, the trace continues
> on the right side of the helix, while the actual ligand path develops
> on the the left of the helix. The trajectory path describes
> more-or-less correctly how the ligand spends a lot of time trapped in
> a potential well, and the direction of exit is correctly indicated by
> the blue trace, but there is a shift with respect to the actual path.
> In this case a serious discrepancy with respect to proximity to the
> helix residues. All that neatly detected with either ribbon or cartoon
> representations of the protein
> I wonder where I am incorrectly performing. I don't see a log file to
> send for inspection.
> Thanks for help
> francesco pietra
> ---------- Forwarded message ----------
> From: Francesco Pietra <>
> Date: Sun, Mar 7, 2010 at 9:55 AM
> Subject: trajectory_path.tcl queries
> To: vmd <>
> Hi:
> New to trajectory_path.tcl, I dare placing a question. In my usage
> source trajectory_path.tcl
> set myligand [atomselect top "resid for.myligand"]
> trajectory_path $myligand scale (which, grossly scales from red to
> yellow, cyan and finally blue)
> there is a large shift in the coordinate scale between the path that
> can be followed from VMD Main and that described with
> trajectory_path.tcl. Also, although the two are identical, the latter
> is on a much smaller scale. The shift is so large that the path
> described with trajectory_path.tcl is seen entirely at the immediate
> outside of the protein.
> thanks
> francesco pietra

Dr. Axel Kohlmeyer
Institute for Computational Molecular Science
Temple University, Philadelphia PA, USA.