From: Axel Kohlmeyer (
Date: Thu Jul 26 2007 - 11:35:46 CDT

On Wed, 25 Jul 2007, Robert Wohlhueter wrote:


RW> Given the definition of the atomselect keyword "within <num> of
RW> <sel>", I would have expected that "<sel>" might accept any valid
RW> selection, including those from other (loaded) molecules: e.g. set
RW> s1 [atomselect 0 "within 5 of [atomselect 1 "all"]"] or set s2
RW> [atomselect 1 "all"]; set s1 [atomselect 0 "within 5 of $s2"]
RW> But such specifications give me "cannot parse atom selection text" errors.

and rightfully so, since the atomselect keyword creates a new function
within the tcl interpreter and that is not part of the selection

RW> Generally speaking, how does one (can one) use the 'within' keyword
RW> to select, say, close contacts between a protein (loaded as one
RW> molecule) and a docked ligand (loaded as another molecule)? How
RW> else might one look at Hbonds between between "all" atoms of the
RW> ligand molecule and a few, close-by protein atoms?

the design of VMD allows selections only from within the same
'molecule' (i.e. atoms loaded from the same file). any other
mechanism would make the outcome of analysis scripts completely
unpredictable, as it would depend on what else you have loaded
into the same vmd session or what you have done before. please
keep in mind that VMD is meant to be a very versatile and flexible
tool and not only designed for the one problem that a specific
person is facing. in that spirit, consistency is a major issue
and many people and particularly john have been working very
hard over the last years to maintain and improve it.

as already suggested, the only way to handle your problem is
to merge the files and then run your analysis (usually not
a big problem, right?).


RW> Bob Wohlhueter
RW> ---------------------------------
RW> Park yourself in front of a world of choices in alternative vehicles.
RW> Visit the Yahoo! Auto Green Center.

Axel Kohlmeyer
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
If you make something idiot-proof, the universe creates a better idiot.