VMD-L Mailing List

From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Fri Apr 29 2011 - 08:36:01 CDT

On Fri, Apr 29, 2011 at 8:48 AM, DreamCatcher
<huangshuping1987_at_gmail.com> wrote:
> Dear vmd-developers,

dear huang,

> I am recently using LAMMPS to do some Molecular dynamics simulation. I have
> two questions to post here.

> First, when I use version 1.9 of vmd to visualize the dump file from the
> simulation. Unfortunately, THE BONDING information is a little wierd as the
> end-point hydrogens which connect to the same carbon form bonds between each
> other. I have looked through the documents  related to topotools trying to
> figure out a solution, however, failed. Nothing about post-processing a
> trajectory files with TCL scripts has been found. Is there any other scripts
> I can resort to to accomplish such a post-processing work? Or is there
> something I missed?

first off, you have to realize two things.

1) lammps dump files do not contain any bonding information or element
information,
    so VMD tries to guess and without knowing the elements, VMD
guesses often badly.

2) VMD was initially written for classical MD problems with a fixed
bond topology and
   thus - for efficiency reasons - it reads (or computes) the bonding
information only once
   and then keeps using that information without regard for what is happening.

this does not mean, that you cannot visualize your trajectory with VMD,
but that it will take some extra effort. people visualizing quantum chemical
calculations (or QM/MM) are facing the same problem and a number of
ways to address this have been found. however, without know any details
and having example files, it is impossible to give specific advice.

> Second, When I do my MD simulation with reax/c codes, the trajectory outputs
> are a little different from those usual dump files due to the extra control

this doesn't make any sense. lammps dump files have a standard format,
which is fully supported. that part of the code is fully independent of the
pair_style you are using.

you have to provide a (small) example file that does not work.

> file designated before the MD run. And I have had some bond and angle
> informations dump to the trajectory files. When I use vmd to view the
> trajectory, I failed. VMD 1.9 cannot recognize the format. What can I do to
> resolve such a problem?

if you want VMD to read a file that is of an unsupported format, you'll have
to write a molfile plugin for it (or some other reader).

cheers,
   axel.

>        Any reply will be appreciated. Thank you in advance!
> Cheers,
> -Shuping Huang 

-- 
Dr. Axel Kohlmeyer
akohlmey_at_gmail.com  http://goo.gl/1wk0
Institute for Computational Molecular Science
Temple University, Philadelphia PA, USA.