Help: Lammps Simulations

LAMMPS LAMMPS Mailing List Mirror
1

Good Day Prof. Karl Hammond,

I am Kumar, a graduate student using Lammps for my research work. In fact, you had helped me during the installation of Lammps parallel version.

I am trying to simulate DNA damage caused by a number of free radicals, and dumped the file in .xyz. I was wondering about any post processing tool or any scripts to count the number of damages caused to Bases- A, G, C, T, sugar molecule, backbone, etc.

Thanks for your help in advance.

With Regards,
Kumar.

2

(I realize this email was addressed to Professor Hammond. Please
excuse the intrusion.)

I doubt there is a generic script for doing this in LAMMPS. We would
need to know more about how you are modelling DNA and what kind of
damage you are considering in your model.

If you are using "pair_style airebo" or "pair_style reax", or a
similar pair-style then I regret I have no idea.

HOWEVER (I posted this response because) if you use "fix bond/create"
or "fix bond/break" to break bonds in DNA, then here is (I just found)
a useful example how to keep track of this information in a "dump"
file:

http://lammps.sandia.gov/doc/compute_bond_local.html

The dump file created in this example contains a record of which bonds
are created and broken. You would still have to write your own short
script or program to read this file, and figure out which bonds were
broken. Try running the simulation, and open up the dump file it
creates (to figure out the format of the headers for each frame).

If this is not what you are doing, then please disregard this email.

Cheers!
Andrew

3

Hello Andrew Jewett,

Thanks for your response. I am using “pair_style reax” for my simulations. I am trying to model the backbone and base damages caused by the free radicals.

  1. If you think this is not straight forward, is there a way that I can figure out the conformational/geometry changes caused due to OH species to the DNA molecule through any plots and probably try to analyze them?

With Regards,
Kumar.

4

I don't know, but I think this information should be something you can
infer from the list of bonds in your simulation. Check out the "fix
reax/bonds" command:

http://lammps.sandia.gov/doc/fix_reax_bonds.html

The file format is not explained, so you will either have to try it
and open up the file it creates with a text editor. Alternately you
can figure it out by opening up the source code at
"src/REAX/fix_bond_reax.cpp" and searching for the "fprintf" statement
(near line 125).

(Of coarse, the first step is obviously to visualize your trajectory
of your simulation. I suspect the bonds created by "pair_style reax"
will not be drawn correctly in most visualization programs, but it's
still worth doing to get an idea. Perhaps somebody can recommend a
good visualizer for this kind of simulation.)

Good luck!

Andrew

5

Has ReaxFF ever been used for DNA?

Steve

6

Yes, ReaxFF has been used previously to understand the DNA damage in the presence of OH species… !!!