From: Thomas C. Bishop (bishop_at_tulane.edu)
Date: Fri Nov 21 2008 - 09:11:42 CST
VMD can read the velocity DCD files saved by NAMD!
However it intepretes them as coordinates so what you see for an
equilibrated system is a spherical distribution centered at zero (no net
velocity) and distance from origin is a measure of an atom's velocity.
Instead of showing bonds display all atoms as points and you'll see a
boltzman distribution. It should be easy to spot (after you've trained
The atoms with the greatest distance (highest velocity) are hydrogens,
so the outer shell of this sphere is white, the inner atoms (lowest
velocity) are the heavy ones the shells should appear colored by mass
order oxygen, nitrogen and carbon (heavest being the inner most shell).
The highest density for any color (i.e. mass) should correspond to max.
in boltzman-distribution for that mass. Cute huh?
There should be no directional bias for a completely random system, but
I can well imagine some atoms will exhibit motion in a line or limited
region of space. Simply calculate the principal components for all
different atom selections to see if this is true or not. If all
principal components are equal then there is no bias. if not spherical
then is this due to system or something wrong with your dynamics.
To my knowledge, this is type of graphical analysis of velocities has
never seen fully exploited. I think there's a lot to learn from this.
You can use all the selection and analysis commands you typically use in
VMD but just remember these are velocities not coordinates.
On Fri, 2008-11-21 at 04:45 -0500, Axel Kohlmeyer wrote:
> On Fri, 21 Nov 2008, dipti lele wrote:
> DL> Hello all,
> hello deepti,
> DL> I am very new to NAMD. I want to analyse the results which i have
> DL> obtained after MD run in NAMD. Now I want to analyse Maxwell- Boltzmann
> DL> energy distribution. But I am not able to load .vel file in VMD. I have
> DL> tried using file type as NAMD binary coordinates but it says that there are
> DL> incosistent atoms in .vel file. I have also checked with the number of atoms
> DL> and they are same in .psf as well as .vel file. I dont know why I am getting
> you are getting this error, because VMD does not (yet) support
> reading velocities. this is in the works, though.
> DL> this error. How should I proceed now? Is there something wrong in the MD
> DL> run due to which this error might occur?
> the more interesting question is, how would you have proceeded
> in case VMD could have been able to read the velocities?
> in principle one can hack the molfile plugin readers to read
> velocities as coordinates.
> DL> Thanking you ..
> DL> With Regards,
-- *********************** * Thomas C. Bishop * *Office: 504-862-3370 * *CCS: 504-862-8391 * *Fax: 504-862-8392 * ***********************
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:50:07 CST