Re: [ccp4bb]: How to make a movie

[Jeff Taylor <jstaylor@duke.edu>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

Edward Berry wrote:
> 
> That sounds almost too simple. Are bond lengths/angles preserved during
> the morph? if a Histidine rotated 180* about CHI2, wouldn't the NE1 and
> CE2 atoms pass through each other on the way to their new positions?
> Do any morphing programs preserve bond lengths and angles during the
> morph? Perhaps one could express the structure in terms of dihedral
> angles and interpolate linearly between the starting and final values
> for the dihedral angles. Still in the case of large-scale movements
> you might get one domain passing through another on the way to its
> final position, where in nature it makes a two-stage movement to go
> around.
> 
> ed

Yes it is too simple.  I just make a linear interpolation between the
old and new coordinates.  This doesn't preserve any physical aspects of
the molecule and the intermediate structures can be severely mangled.  A
180 degree rotation of a Histidine (or tyrosine, phenylalanine, etc.)
will look really funny.  To make a physically reasonable morph one would
need to employ some kind of directed molecular dynamics, or at least
incorporate some kind of energy minimization for the intermediate
steps.  What a mess.  On the bright side, the linear interpolations
often work quite well, provided that the motions aren't too large or
complex, AND they aren't taken too seriously.

In the end, while I think they can be useful is some cases, I don't take
molecular movies too seriously in their current form.

- Jeff Taylor
 
-- 
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Dr. Jeffrey S. Taylor       Duke University Medical Center
Department of Biochemistry  Box 3711, Nanaline Duke Bld  
Phone (919) 681-5266        Durham, NC 27710