From: Axel Kohlmeyer (axel.kohlmeyer_at_theochem.ruhr-uni-bochum.de)
Date: Tue Jul 13 2004 - 15:28:17 CDT

On Tue, 13 Jul 2004, Kenneth Smith wrote:

KS> Hello.

hi,

KS> I am currently working on simulations where we allow the bonding
KS> structure to be altered after each time step in the MD simulation. I
KS> would like to use VMD as the visualization program, but so far I can
KS> not identify any possible trajectory file formats that allow for
KS> updated connectivity at every time step. The dynamic bonding method
KS> is not sufficient as we may break specific bonds and also at times
KS> form bonds with radical molecules which would have the possibility of
KS> one additional bond beyond the traditional valance limits. Are there

the dynamic bonds visualization does not in any way account for
valences, it only uses a distance criterion. when combined with
elaborate selections, this can produce rather nice results.
please have a look at:
http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/cpmd-vmd/part4.html#chap6_sect1
which demonstrates how some of these selection are constructed
(although the specific example only covers the selection of full molecules
based on a distance criterion on a single molecule).

to improve that it would be really nice, if there would be a way in VMD to
specify different selections for both ends of the dynamic bond. e.g. if i
want to have dynamic bonds of 'heavy' atoms with hydrogens, then i need to
specify a short bondlength, but that will create many bogus bonds between
the 'heavy' atoms.

the second way would be that you create a lot of psf/pdb files (one for
each frame) and then use an adapted version of the attached script to
load each of them into a separate molecule and display them sequentially
with 'molmovie'. not very elegant, but it works.

a third way would be to read the bonding information into an array
and then use the 'trace' mechanism to apply this information via
an atom selection function and the setbonds subcommand. if there are
only a few selected bonds that change (or that cannot be represented
via dynamic bonds), this could be quite doable.

feel free to get back to me, if you need more help with that.

regards,
        axel kohlmeyer.

KS> any ways to address these issues with VMD as it is currently written
KS> or would this require source code modification?
KS> Thanks for you help! Ken

KS>
KS> -----------------------------------------------------------------------
KS> Kenneth D. Smith
KS>
KS> Department of Chemical Engineering
KS> University of Massachusetts- Amherst
KS> 159 Goessmann Lab
KS> 686 N. Pleasant St.
KS> Amherst, MA 01003
KS> Office: 216 Goessmann
KS> tel: (413) 545-2234
KS> fax: (413) 545-1647
KS> email: ksmith_at_ecs.umass.edu
KS>
KS>

-- 
=======================================================================
Dr. Axel Kohlmeyer                        e-mail: axel.kohlmeyer_at_rub.de
Lehrstuhl fuer Theoretische Chemie          Phone: ++49 (0)234/32-26673
Ruhr-Universitaet Bochum - NC 03/53         Fax:   ++49 (0)234/32-14045
D-44780 Bochum  http://www.theochem.ruhr-uni-bochum.de/~axel.kohlmeyer/
=======================================================================