From: Axel Kohlmeyer (
Date: Tue Jan 20 2009 - 17:21:16 CST

On Tue, 20 Jan 2009, nicholas lee wrote:

NL> Thanks John,

NL> But I still must be doing something wrong. Screenshots of what I am
NL> expecting are shown at the links below.

NL> Here, I have loaded the molecule configuration from the TK console,
NL> entering source->mol-> lmpbondsfromdata,etc. I want the interatomic
NL> bonds break but they do not.

does you lammps data file contain bonding information?
this is what the lmpbondsfromdata script is supposed to
help with (i should know, i wrote it). however just sourcing
the script is not enough you have to call the proper procedure
with the proper arguments. check out the description.

NL> Here, I have simply loaded the molecule configuration by
NL> double-clicking my lammpstrj file. I want to connect the gray atoms
NL> using the "dynamic bonds" representation but they never connect, no
NL> matter how large i set the cut-off bond length.

there are a bunch of possible reasons. the main reason being,
that in lammps all atoms are labeled 1, 2, 3 and so on, but
VMD expects atoms to have names representing the elements.
there are some scripts in the lammps vmd tools directory with
procedures to help in that as well, but you have to do it or
else VMD has no clue.

what units are you using and what distance do you have between
your atoms (can be measured from the GUI by mouse click).
VMD assumes angstroms.

NL> In both cases, when I switch from "CPK" to "DynamicBonds"
NL> representation, my atoms disappear and I get a blank display window
NL> shown at this link.

further debugging would require access to the a piece of the
lammps format trajectory file and the corresponding data file.


NL> Maybe you can have a look at my settings in the
NL> screenshots....Thanks again for your help.

NL> Nicholas J. Lee

NL> More details: From the User's Guide, it seems as if I have made all
NL> the necessary selections in the menu and that there is nothing more
NL> I need to do other than provide my geometry data.

not quite. the lammps data format is withholding some key information
for VMD to guess some information for what it needs. i usually fix up
lammps data from a single lammpstraj frame with the script in the
respective tools directory and then write out a proper .psf format
file that contains all information that VMD needs and then use the .psf
in combination with .dcd output in lammps for visualization.

breaking bonds is not expected by default in VMD, so that requires
overlaying dynamic bonds with a small radius VDW rep, as john already
pointed out.



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.