Restraints and System COM Drift

From: Philip Blood (
Date: Mon Feb 26 2007 - 13:48:01 CST


I am using SMD, with a zero velocity, to restrain the COM of a protein
along a single coordinate direction. The protein is embedded in a lipid
bilayer. Since I am using langevin dynamics (and also SMD) the momentum
of the system is not conserved, and I observe center of mass drift of
the entire system. However, since I have restraints on my protein, it
does not drift along with the rest of the system. This results in a
situation where the lipids are "tugging" on the protein as they move due
to COM motion of the entire system, but the protein resists this motion
since it is restrained to reference coordinates that (of course) do not
move with the system COM drift. When the protein is not restrained, the
COM drift does not present a problem since the protein just goes along
for the ride with everything else. But in this case it creates a
situation which is unphysical.

Has anyone else experienced this problem? Any suggestions on how to
work around it? I recognize that NAMD has an option now to remove COM
motion every (electrostatics) timestep, but, as pointed out in the
manual, I would be leery of using this option with langevin dynamics and
SMD. I suppose I could use zero velocity SMD to restrain the COM of the
entire system, and then use selectConstraints to constrain certain atoms
of my protein along the desired coordinate direction (although I would
lose the ability to use the SMD COM restraint for the protein). I am
not sure if this would be any better than just using the zeroMomentum
option. Any insight is appreciated.


This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:44:25 CST