From: John Stone (
Date: Mon Jan 23 2012 - 15:24:51 CST

If you use a file naming convention that guarantees that lexical sorting
will result in the right order of files, you could simply use the Tcl
"glob" and "sort" commands in VMD, thereby generating the command line
to launch catdcd, either by "exec" or other means.

  John Stone

On Mon, Jan 23, 2012 at 03:11:34PM -0600, Bryan Roessler wrote:
> Hi,
> I'm writing an automated analysis script that I want to be able to run on
> all of my NAMD output, regardless of directory or file names. I've been
> able to get most of it written except for the very important first step
> where I concatenate all of my dcd output files into a single file, whose
> output name I can control (and therefore run the analysis). I was
> wondering if there was a way to load all dcd files within a directory into
> CatDCD.
> My current example:
> set name "Project Name"
> catdcd -o protein_only.dcd -otype dcd -i protein_only.ind
> ../$name_sim1.dcd ../$name_sim2.dcd ../$name_sim3.dcd ../$name_sim4.dcd
> ../$name_sim5.dcd ../$name_sim6.dcd ../$name_sim7.dcd ../$name_sim8.dcd
> ../$name_sim9.dcd ../$name_sim10.dcd ../$name_sim11.dcd ../$name_sim12.dcd
> ../$name_sim13.dcd ../$name_sim14.dcd ../$name_sim15.dcd
> ../$name_sim16.dcd
> If I run additional simulations in the future (sim17, sim18, etc), I want
> to be able to rerun the analysis after concatenating the dcd's without
> having to manually edit the analysis script.
> Is there some way in Tcl or VMD that I can call all dcd files in a
> directory?
> I guess the messy option would be to add a huge number of dcd entries into
> the above script and allow CatDCD to fail to find them if they had not yet
> been created.
> Thanks,
> Bryan Roessler

NIH Resource for Macromolecular Modeling and Bioinformatics
Beckman Institute for Advanced Science and Technology
University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801           Phone: 217-244-3349       Fax: 217-244-6078