Re: Calculating displacement in a system with periodic boundary

From: Joshua Adelman (
Date: Mon May 11 2009 - 23:44:23 CDT

If you had perfect knowledge of every atom at every time step of the
simulation I imagine that you could could calculate the path length by
checking for the case when a particle goes through the boundary and
comes in the opposite side. The other option is to re-run the
simulation and don't wrap the coordinates. It will look like your
system is drifting apart when you look at the dcd, but the PBC should
be applied when NAMD calculates the forces. You could then post-
production re-wrap the coordinates using the PBC tools in VMD.


On May 11, 2009, at 9:26 PM, Fred (Rui FENG) wrote:

> Hi All,
> I have a question that can be simplified like this:
> Suppose there's a cube with lots of particles/atoms. The boundary is
> periodic and only Lennard Jones force field is in the system. All
> atoms are moving with random velocities. The velocity of the center
> of mass could not be zero. I have a common DCD file with coordinates
> of all atoms. My question is: how to get the displacement of all
> atoms?
> I cannot just simply subtract the coordinate of one atom from the
> coordinate of the same atom in the previous frame. Supposing the
> cube has 200 edge length and the center is (0,0,0). If a atom at the
> position (99, 0, 0) moves to the position (-99, 0, 0), the
> displacement could be either 2 or -198 because of the periodic
> boundary. The absolute coordinate could be (102, 0, 0), but in the
> DCD file it's (-99, 0, 0). It makes the displacement calculation
> impossible.
> So I was wondering if there are any ways with which NAMD can output
> a "modified" DCD file with absolute coordinates. Thank you!
> Have a nice day!
> Fred

Joshua L. Adelman
Biophysics Graduate Group Lab: 510.643.2159
218 Wellman Hall Fax: 510.642.7428
University of California, Berkeley
Berkeley, CA 94720 USA

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:52:46 CST