[lammps-users] Visualize by VMD but no showing any bonds


I am using VMD 1.8.7 to visualize the dump file from lammps by creating a psf file first, but there is no bond but only atoms information.

Okay, the command I ran were simply following the lmp2vmd manual, such as the micelle example provided by lammps:

source lmpbonds2vmd.tcl

mol new dump.micelle type lammpstrjwaitfor all

lmpbondsfromdata 0 data.micelle

animate write psf micelle.psf.

However, the screen log shows that:

Atoms: 1200

Bonds: 0

And the VMD only shows the points of atoms with no bonds as well.

Is there anything I forgot?

Thank you,



the script you are using is obsolete and has been
superseded by the topotools plugin, which has many,
many more features and is far more flexible.

the following commands (text mode or gui mode)
will generate a suitable .psf from a data file.

package require topotools
topo readlammpsdata data.micelle bond
animate write psf micelle.psf




please keep the list in cc:
i've stopped responding to personal e-mails that
don't concern me personally. this case is an exception,
since it concerns software that i have written.

Thank you for your quick response. I did that. But there is still some lines
across the cube in this visualization as in the attachment. Do you know what
is that? And how to import the trajectory file into this data file?

those are bonds that cross the periodic boundaries.
the show up since the data file does not have any
image flags set. this does not affect the calculation,
but the visualization.

under normal circumstances, this should be fixable
by using:

package require pbctools
pbc join res
pbc unwrap -all
pbc wrap -compound res -all

but in this case it fails at the pbc join step,
probably due to some heuristics in the pbctools
plugin that don't apply here.

furthermore, the lammps trajectory reader does
reorder atoms by atom tags to be consistent
across multiple files, however that will produce
a different order than in the data file. i have
to think about an option or tool that will allow
to reorder atoms while keeping their bond, angle,
and other definitions intact.

the easiest way to deal with it right now is to
write out the trajectory in .dcd format instead
which will preserve the order of the atoms.
this still does not fix the bonds across the
boundaries issue, though.

if you generate a "normal" data file,
this should not be a problem.