RE: Periodic Boundary Conditions

From: JC Gumbart (
Date: Mon Mar 26 2007 - 09:42:15 CDT

I personally always set the center to be the center for everything, not just
the protein. It's presumed you centered the protein in your water box to
start with anyway so maybe this won't make too large a difference.

Is this a membrane system? If so, I would also suggest basing the x and y
basis vectors on the water dimensions, not the lipids, which stick out
beyond the periodic boundary.

Otherwise, what you see is not wrong. If you don't wrap the coordinates,
when displayed, they will appear to be diffusing out (useful for calculating
various physical properties). In NAMD though, they are still within the
same periodic box.

-----Original Message-----
From: [] On Behalf
Of Per Jr. Greisen
Sent: Monday, March 26, 2007 2:26 AM
Subject: namd-l: Periodic Boundary Conditions


I have applied periodic boundary condition to a md simulation. I use the
following commands in order to get the values for the PBC

set everyone [atomselect top all]
measure minmax $everyone
measure center $everyone

{-40.5429992676 -40.0960006714 -55.2290000916}
{23.1319999695 25.5699996948 53.1780014038}

and with the center of the protein as
-11.4283914566 -9.76103401184 -0.764146029949

I set the parameter
cellOrigin -11.43 -9.76 -0.76

and the basis vector as the nummerical sum
cellBasisVector1 63.67

When I display the md simulation - it start out nice as I expect with a
square box but as it continues the waters starts diffusing out. So my
question is - the MD simulation uses the periodic boundary condition but
the vizualisation looks change and can be fixed with the

wrapWater on

Any help or advice appreciated. Thanks in advance

Best Regards
Per Jr. Greisen

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:46:07 CST