Ion Channel Jarzynski

From: Alessandro Cembran (
Date: Thu Aug 10 2006 - 17:56:29 CDT


I am trying to compute the PMF for the passage of a molecule through a
protein channel by using the Jarzynski equality (the constant velocity
pulling through a dummy atom scheme). I've read several papers on this
topic, but I couldn't find the fine details on how to set up the
constraints. I also ran the deca-alanine, ubiquitine and nanotubes
tutorials to get acquainted with the methods but I still have some
things that need to be clarified:

- I need to pull down the channel the molecule through the dummy atom
that moves at a constant velocity, and I can do that with a tcl script.
But this is not enough, I need a second handle to keep the protein from
moving (I realized that in 4ns of simulation the protein and the
membrane can move several angstroms along the z direction, as somebody
else noticed previously on this forum). In the deca-alanine and
ubiquitine systems this second handle was easily provided by fixing an
atom, but in this case it is not possible. It is the center of mass of
the protein that should be fixed, at least along the z direction and I
don't think that is possible. Would it be meaningful to add a
restraining force to the COM of the protein through a tcl script? If
yes, how would this additional fluctuating force affect the computed PMF?

- In this article (Amaro,Tajkhorshid,Luthey-Schulten - Proc. Natl. Acad.
Sci. USA 2003, 100(13), 7599.) I read that there is no substantial
difference in computing a PMF by Jarzynski equality with a NPT or NVT
simulations. I was wondering if anybody else has any comments about that.

Thanks in advance,


Alessandro Cembran,PhD
Post Doctoral Associate
Univ. of Minnesota, Dept. of Chemistry
G2, 139 Smith Hall 207 Pleasant St SE
Minneapolis, MN 55455-0431
Office: +1 612-625-4319

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:42:28 CST