# Re: center of mass motion: problems

From: Leandro Martínez (leandromartinez98_at_gmail.com)
Date: Thu Dec 08 2005 - 17:49:23 CST

It depends of what you want to compute. If you want to compute
the rmsd of a protein, in general the protein coordinates of each frame
are fit to the initial, or average, coordinates. If you want to compute
diffusion
coefficients, you shouldn't wrap or move your coordinates at all.
Anyway, what you mean by center of mass movements at least
initially? The center of mass of the whole system
probably oscillates slightly from
frame to frame. The center of mass of a protein certainly will move in
space, and that is what is related to the diffusion of the protein.
If you have a center of mass drift in the first stages
of your simulation and not thereafter, probably you should do some
equilibration.
Leandro.

On 12/8/05, Vani Krishna <vakri2002_at_yahoo.com> wrote:
>
> Hi all,
>
> I know this question has been asked before. But Iam
> still a bit confused. I always observe a center of
> mass motion drift if I do not wrap my system, atleast
> initially. If a have such a trajectory and would like
> to do analysis, I wanted to knwo the best way to go
>
> (1) fitting all the frames wrt the initial frame and
> then do the analysis
>
> (2) move all the frames so that they are centered (tcl
> below) and then do the analysis.
>
> are there any suggestions out there? do people do
> anything else to handle this problem?
>
> Thanks
> -------------------------------------------------------------
> proc moveorigin {text} {
> set sel [atomselect top \$text]
> set outfile stdout
> set out stdout
> set nf [molinfo [\$sel molid] get numframes]
> puts "number of frames : \$nf"
>
> set frame 0
> for {set i 0 } { \$i < \$nf } {incr i } {
> \$sel frame \$i
> \$sel update
> set shift [measure center \$sel]
> set c1 [lindex \$shift 0]
> set c2 [lindex \$shift 1]
> set c3 [lindex \$shift 2]
>
> set c1 [expr -\$c1]
> set c2 [expr -\$c2]
> set c3 [expr -\$c3]
>
> set shiftnew "\$c1 \$c2 \$c3"
> \$sel moveby \$shiftnew
>
> }
>
> }
>
>
> __________________________________________________
> Do You Yahoo!?
> Tired of spam? Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:41:25 CST