From: Thomas C. Bishop (bishop_at_tulane.edu)
Date: Tue Jul 20 2010 - 16:51:17 CDT

Cool! that works.... the animation slider boounces back and forth as the
animation works it's way across the trajectory...b/c the RMSD more or less
increases monotonically during my sim.

This is a quick way to make any sim look like a directed protein folding
simulation but that's cheating :-)

Now on to more sophisticated gotolist selections...
1) rmsd of a flexible loop
2) some angle/conformation of interest
3) a list of interaction energies
4) the distance of some water molecule of interest
5)the binding of a ligand
the list is endless

All I have to do is feed your loop the gotolist!
Slick!

Thanks
Tom

On Tuesday 20 July 2010 01:04:57 pm Axel Kohlmeyer wrote:
> tom,
>
> On Tue, Jul 20, 2010 at 1:17 PM, Thomas C. Bishop <bishop_at_tulane.edu> wrote:
> > Does anyone have a tool handy that allows you to arbitrarily reorganize
> > the animation display order of a DCD file.
>
> no. you'd have to swap coordinates around if you want to use
> the standard animation control dialog.
>
> > For example play a trajectory in the order of increasing RMSD instead of
> > the actual time order.
> >
> > This can be accomplished with a list of frame numbers and an "animate
> > goto" coupled with a trace variable in tcl.
> > In general the list of frame numbers can come from anywhere.
>
> this would not work, since the animate goto would trigger
> the reorder event and you'd have a deadlock.
>
> the best you can do is something like the following (untested):
>
> set mol [molinfo top]
> set nf [molinfo $mol get numframes]
> set sel [atomselect $mol all]
> set refsel [atomselect $mol all frame 0]
> set rmsdlist {}
>
> for {set i 0} {$i < $nf} {incr i} {
> $sel frame $i
> lappend rmsdlist [list $i [measure rmsd $sel $refsel]]
> }
>
> set gotolist [lsort -increasing -real -index 1 $rmsdlist]
>
> for {set j 0} {$j < 10} {incr j} {
> for {set i 0} {$i < $nf} {incr i} {
> animate goto [lindex $gotolist $i 0]
> display update
> display update ui
> after 100
> }
> }
>
>
> if you want something more fancy with a GUI,
> you could rip off the multimonanim plugin for it.
>
> cheers,
> axel.
>
> > The problem however is what should the trace variable be... I guess it
> > can't be the frame number.
> >
> >
> > Any tips or suggestions much appreciated.
> > Thanks in advance,
> > TOm
> >
> > *******************************
> > Thomas C. Bishop
> > Tel: 504-862-3370
> > Fax: 504-862-8392
> > http://dna.ccs.tulane.edu
> > ********************************
>

*******************************
   Thomas C. Bishop
    Tel: 504-862-3370
    Fax: 504-862-8392
http://dna.ccs.tulane.edu
********************************