VMD-L Mailing List

From: Jérôme Hénin (jerome.henin_at_ibpc.fr)
Date: Thu Apr 04 2013 - 08:31:49 CDT

Hi Tom,

Don't tell me you don't have a biocore account!!!
Okay, I didn't want to clutter the mailing list archives with attachments, but this so tiny it shouldn't be a problem.

Cheers,
Jerome

----- Original Message -----
> Jerome,
> I'd like to download/install but need a biocore login.
> did you intend to have this password protected on the biocore site?
> a quick url for obtaining uname/pword will save me from looking.
>
> I hope qwrap works as advertised I'm in need of such a tool!
>
> Tom
>
>
>
> On 04/03/2013 04:02 PM, John Stone wrote:
> >
> > Thanks Jerome. I will take a look at this when I have a chance.
> > One of the things I've wanted to do is to build in a very fast
> > wrapping helper routine into VMD itself that would be useful to
> > speed up pbctools and to use for some of the internal calculations
> > that VMD does itself, which currently use hand-coded and/or
> > fully inlined PBC handling code. When I get a chance to dig into
> > it, I'll send you a ping. I'd love for other people to chime in
> > too, since this is an area of VMD that generally needs some work.
> >
> > Cheers,
> > John Stone
> > vmd_at_ks.uiuc.edu
> >
> > On Tue, Apr 02, 2013 at 05:05:11PM +0200, Jérôme Hénin wrote:
> >> Hi everyone,
> >>
> >> I'd like to mention a tool for people who often wrap large MD
> >> trajectories in PBC, and are impatient like me. It's a Tcl proc
> >> implemented in C++ and called qwrap.
> >>
> >> https://biocore-s.ks.uiuc.edu/biocore/biofs/VMD%20(Public)/src/jerome/qwrap.tar.gz
> >>
> >> Instructions:
> >> $ tar xf qwrap.tar.gz (in own directory)
> >> $ make
> >>
> >> In VMD:
> >>> load path/to/qwrap.so
> >>> qwrap [first n] [last n] [compound none|res|beta]
> >>
> >> Compared with PBCTools it's a coarse and rigid tool, mostly built
> >> to answer my own needs (e.g. orthorhombic cells only, the center
> >> of the cell is (0,0,0)...). But in my hands, it is 10 to 30 times
> >> faster.
> >>
> >> The compound-based wrapping doesn't use a specific reference atom,
> >> but the center of the wrapping group. Which btw can be defined
> >> either as residues, or by setting custom flags in the beta field,
> >> so you can keep a macromolecule in one piece. The flags should be
> >> integers, and could be alternating 0s and 1s or anything else,
> >> they only need to change when the next atom belongs to a
> >> different wrapping group.
> >>
> >> A more mature version of this could eventually provide a fast
> >> back-end for the pure-Tcl pbc wrap.
> >>
> >> It's work in progress, with the usual caveats as to bugs etc.
> >> Please let me know if it is useful, and if you improve it, share
> >> it back! Here, let's say I release it under the GPL.
> >>
> >> Cheers,
> >> Jerome
> >
>