From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Wed Jul 30 2008 - 23:45:25 CDT
if you're willing to alter your approach a little bit, since namd has a
complete embedded tcl interpreter you can run namd until the first step
of interest, perform whatever processing you want (including calling a
perl script or vmd) *within the namd config file* on the most recently
written restart file, and then continue the namd run without restarting.
This action could, of course, include stripping the file of solvent or
fully analyzing it.
> I am running a script in perl that sets up a protein-pigment structure,
> solvates in a box, adds ions, and then runs a NAMD min/equilibration
> followed by calculations on the ouput structure file. Currently, I am
> outputting restart files with appended step #-labels so that I can perform
> subsequent calculations on the structure for once every few steps.
> However, my output is then rather large by the time NAMD is done (several
> GB) and for my purposes this can't be allowed. I would like to either be
> able to ouput smaller files or possibly perform calculations while NAMD is
> running so the files already read can be deleted.
> My question, then, is how can I output just the pigment-protein
> coordinates each time, is there something better than using the restart
> files? My only thought so far would be to run NAMD till the first
> step-of-interest and then output a .coor file and use VMD with another tcl
> script to get rid of the solvation box. However, if there is a way to do
> this without repeatedly restarting NAMD, this would be best because my
> perl program is presently waiting for NAMD to finish entirely before
> Thank you,
> University of California, Berkeley
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:48:06 CST