Re: question about rotation force in SMD

From: Giacomo Fiorin (giacomo.fiorin_at_gmail.com)
Date: Wed Apr 27 2011 - 10:00:33 CDT

Hello Jianing, I think you could use two options to force the conformational
change to happen in a short time. One is to use targeted MD, which is
specifically designed to drive a system towards a set of final reference
coordinates (only forcing the heavy atoms, or even only the backbone would
be better). Another is to use the collective variables module: a starting
point could be to define a variable such as the RMSD wrt the final
coordinates, and apply a moving harmonic restraints (thereby emulating
targeted MD). But you can also define and monitor many other variables of
your choice, identify those that best describe the transition, and then
force them.

Giacomo

---- ----
  Giacomo Fiorin - Postdoctoral Researcher
  ICMS - Institute for Computational Molecular Science - Temple University
  1900 N 12 th Street, Philadelphia, PA 19122
  giacomo.fiorin_at_temple.edu
---- ----

On Wed, Apr 27, 2011 at 10:34 AM, jnsong <jnsong_at_itcs.ecnu.edu.cn> wrote:

> Dear all,
>
> Recently, I want to use SMD to research the mechanism about how a
> membrane transporter transport its substrate through the lipid bilayer.
>
> The transporter is a transport membrane protein and has six helix to
> compact to be a cylinder in the lipid bilayer.
>
> I suppose a mechanism that the transporter will rotate around the
> bilayer normal at the same time the six helix change to an open state to
> allow the substrate to travel through the channel.
>
> So I have added a rotation force to allow the transporter to rotate
> around an axis parallel to bilayer normal as told by the NAMD force
> tutorial.
>
> But the result is a little disappointing. The transporter just rotate
> around the axis parallel to bilayer normal, but doesn't change to an
> open state to allow the substrate to travel.
>
> Would you someone please tell me what can I do to improve the result?
>
> Should I increase the angular acceleration ongoing during the whole
> simulation rather than keep it a constant value?
> what is a reasonable value of angular acceleration?
>
> The value of angular acceleration in my simulation is 1 rad.ps-1.
> My whole simulation time is 5ns.
>
> Thank you very much!
>
> Jianing
> --
> Jianing Song
> Ph. D. Student
> Institute of Theoretical and Computational Sciences
> Dept. Phys.
> East China Normal University
> 200062
> 3663 North Zhongshan Road
> Shanghai, P. R. China
>
>

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2012 - 23:20:11 CST