[MMTK] DCD to XYZ conversion

konrad.hinsen at laposte.net konrad.hinsen at laposte.net
Tue Dec 14 12:03:44 CET 2004


On Dec 13, 2004, at 17:17, Kaihsu Tai wrote:

> After trying a few things, it seems the most reliable way is
> to convert the DCD files into XYZ format one frame per file,
> and then read them back in (we are doing exactly this for
> Gromacs).
>
> Does anybody have an API in the C language (or Python) that
> can do something like this?  (Web searches have been
> fruitless.)
>
> http://bugs.biosimgrid.org/show_bug.cgi?id=176
>
It seems your problem is reading the DCD files. I know this is not so 
trivial as there are many variants. MMTK uses a DCD parser from VMD 
(somewhat modified), which has always worked fine for the few DCD files 
I have dealt with. I know it doesn't deal well with some DCD files 
produced by recent versions of CHARMM, but I have neither such files 
nor a description of the latest format variants.

I suspect that recent VMD versions do a better job of reading DCD than 
MMTK does, so Justin's proposition to use VMD for conversion is perhaps 
a good one. MMTK would benefit from an update of its DCD reading code, 
but it's not something I expect to do any time soon, given time 
constraints and the extreme unpleasantness of the task.

My personal take on this is that ill-defined and badly documented file 
formats such as DCD are not a good choice for databases. I know that 
CHARMM is a very popular program, but then BioSimGrid is an important 
project as well. If anyone has the weight to make the CHARMM people 
implement and document a decent trajectory format, it's you!

Konrad.
--
---------------------------------------------------------------------
Konrad Hinsen
Laboratoire Léon Brillouin, CEA Saclay,
91191 Gif-sur-Yvette Cedex, France
Tel.: +33-1 69 08 79 25
Fax: +33-1 69 08 82 61
E-Mail: hinsen at llb.saclay.cea.fr
---------------------------------------------------------------------




More information about the mmtk mailing list