Re: Alignment of velocity dcd file

From: Axel Kohlmeyer (akohlmey_at_gmail.com)
Date: Sat Aug 22 2009 - 15:49:19 CDT

On Sat, 2009-08-22 at 23:03 +0300, Mert Gür wrote:
> Dear all,
> Lets say that I have a coordinate dcd file "dcd1" and a velocity dcd
> file "vel1" both of size 2000x1200.
> I align my protein using vmd and save the aligned coordinates as
> "dcd2".
> Now I can easily obtain an rotation matrix rot of size 2000x2000 that
> performes the following transition
>
> rotxdcd1=dcd2.
>
> What I want to do is to eliminate the terms in the velocity due to
> transition and rotation of my protein. Does the following transition
> make sense in that light?

no. when you re-align the coordinates your rotation matrix
contains the _accumulated_ rotation and translation. but
the velocities file contains the _instantaneous_ velocities.
to "purge" the velocities, you would have to subtract the
center of mass velocity and the rotational velocities around
the three moments of inertia.

however, you have to think hard, if this is really what
you want. with this procedure, you also remove the physical
fluctuations. i would thus first do a check if you get
an averaged drift and how much this contributes to the
total energy. rotation is practically negligible in typical
aqueous bulk systems.

cheers,
   axel.

> rotxvel1
>
> I am asking because the velocities seem to increase if I do this
> matrix multiplication. Since I am eliminating some of its components,
> the velocities should actually decrease.
> So at that point I started to doubt if the velocities due to
> transition and rotation are included in the velocity dcd file?
>
> The manual just says
>
> Description: The binary DCD velocity trajectory filename. This file
> stores the trajectory of all atom velocities using the same format
> (binary DCD) as X-PLOR. If velDCDfile is defined, then velDCDfreq must
> also be defined.
>
> Can anybody help me out with that?
>
> Best,
> Mert
>
>
>

-- 
Dr. Axel Kohlmeyer  akohlmey_at_gmail.com 
Research Associate Professor
Institute for Computational Molecular Science
College of Science and Technology
Temple University, Philadelphia PA, USA.

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:53:12 CST