From: John Stone (johns_at_ks.uiuc.edu)
Date: Wed Feb 23 2011 - 15:45:48 CST

Francesco,
  I am working on this, but to implement it well is far more complex
than one might guess looking at the Tcl script implementations. My goal
is to make it work both for interactive sessions (visualization+analysis)
as well as batch mode scripting (ala BigDCD). This is a feature that
will start showing up in normal VMD builds sometime later this year.

The trickiest aspect of this is to make it usable for scenarios where
one does analyses that can potentially require modifying the
coordinates (e.g. alignining structures as a first step) or other
per-timestep data, which requires writing changes back to disk, so they
can be restored when the trajectory frame is referenced again.

Anyway, this particular feature is VERY high on my TODO list...

Cheers,
  John

On Wed, Feb 23, 2011 at 10:37:34PM +0100, Francesco Oteri wrote:
> Dear VMD developers,
> I attempted to analyze a MD trajectory with vmd but I've found two big
> challenges:
> 1) VMD loads the entire trajectory and this causes a great time wasting
> 2) If the trajectory is too large, because of the memory usage, the PC
> hangs and when it reaches the 85% VMD is killed by the OS.
>
>
> I think that should be implemented an option that permit to load a
> trajectory frame by frame.
> In particular I think that vmd should load the first frame of the
> trajectory and each time the informations about a frame are required,
> the data should be loaded from the disk.
>
> I know that the script "bigdcd" permits to bypass the problem but, in my
> opinion, a native implementation of this feature should make easier to
> use the MD analysis tools already present in VMD.
>
>
> Francesco

-- 
NIH Resource for Macromolecular Modeling and Bioinformatics
Beckman Institute for Advanced Science and Technology
University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
Email: johns_at_ks.uiuc.edu                 Phone: 217-244-3349
  WWW: http://www.ks.uiuc.edu/~johns/      Fax: 217-244-6078