Re: Importing a DCD into Mathematica

From: Joshua Adelman (jla65_at_pitt.edu)
Date: Wed Apr 11 2012 - 07:36:07 CDT

Hi Ajasja,

If you are just looking to avoid writing a lot of analysis in TCL, and want to code in a language that is more amenable to mathematical manipulations, you might consider looking at the following two projects written in Python:
MDAnalysis - http://code.google.com/p/mdanalysis/
Prody - http://www.csb.pitt.edu/prody/index.html

They both can read dcd formatted trajectories and then you can use python's numerical and science libraries which are very robust (numpy/scipy). There are actually a lot of molfile readers floating around the web wrapped for python (e.g. DE Shaw's Timescapes package or Desmond itself).

Personally I use VMD for analysis if I don't have to write much additional TCL, but if I'm doing a novel analysis with any sort of complicated mathematical manipulations, I almost always use a molfile reader in Python and a custom script. I've had a lot of success with this strategy in combination with Cython (http://cython.org/) if I ever need to speed up slow code that one can't vectorize using Numpy.

Best wishes,
Josh

On Apr 11, 2012, at 8:20 AM, Ajasja Ljubetič wrote:

>
> not that i don't see your point (and a reasonable one), but
> you _could_ make the same argument for adding those
> features to VMD/Tcl.
>
> Yes, but I don't think I'll be able to port all the features of Mathematica to tcl:)
>
> also, it always pays to do a trip to the Tk/Tcl wiki.
> a *lot* of solutions can be found there for pretty
> usual and fairly unusual problems.
>
>
> I did find a nice Map implementation a while ago, and one could probably use lambda functions as well, but for me it feels a bit hackish.
>
> what might be an interesting option would be
> to write a C interface/import into mathematica for the
> VMD plugin interface. that would be less tedious *and*
> in addition you would gain compatibility to *all* supported
> file formats.
>
>
> Hmm, that *is* an interesting idea. I've never need to interface C and mathematica, but I'll look into it.. Starting with CatDCD.c
>
> Best regard,
> Ajasja

--
Dr. Joshua L. Adelman
Department of Biological Sciences
University of Pittsburgh, Pittsburgh, PA

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2012 - 23:21:26 CST