[MMTK] NMA

Konrad Hinsen konrad.hinsen at cea.fr
Fri Mar 17 14:42:45 CET 2006


On Mar 17, 2006, at 13:59, vanitha at cs.wisc.edu wrote:

> Can I try the following approach:-
>
> 1) Create Universe with Deformation Force field
> 2) Add unbound structure 1.
> 3) Compute the normal mode for it and store the normal mode vectors  
> in an
> array.
> 4) Remove unbound structure 1.
> 5) Add unbound structure 2 and repeat steps 3 & 4 for structure 2.
> 6) Set Force field to the all-atom Amber94ForceField.
> 7) Add the complex containing structure 1 and 2.
> 8) Perform refinements using the normal mode vectors stored in the  
> arrays.
>
> Does this make sense? I'm not sure how to go about doing what you
> suggested and that's why I'm looking at this approach.

No, that won't work. As soon as you add or remove molecules, the  
universe becomes for all practical purposes a different one. This  
means that for example the  normal modes computed in your step 3  
won't be compatible with the universe after step 4.

Here is my suggestion:

1) Put the complex into the universe.
2) Make a copy of the configuration:
		complex_configuration = copy(universe.configuration())
3) Move one of the ligands away:
		ligand1.translateBy(Vector(10., 0., 0.))
    10 nm is sufficient for a deformation force field, but feel free  
to use more.
4) Add the deformation force field:
		universe.setForceField(DeformationForceField(cutoff=2.5))
5) Calculate the normal modes:
		modes = EnergeticModes(universe)
6) Restore the complex configuration:
		universe.setConfiguration(complex_configuration)
7) Set the Amber force field:
		universe.setForceField(Amber99ForceField(...))
8) Do refinement using the normal mode vectors.

A comment on the normal modes (step 5): Since the two molecules do  
not interact, you will have a total of 12 zero-energy modes (the  
translation and rotation of the two molecules). Any other mode will  
contain a deformation of only one of the two molecules, except if the  
two molecules are identical (in that case there are degenerate  
modes). So even though you calculate the modes for the two molecules  
together, you can still separate them into deformations of the  
individual molecules, which I think is what you want.

Konrad.
--
---------------------------------------------------------------------
Konrad Hinsen
Laboratoire Léon Brillouin, CEA Saclay,
91191 Gif-sur-Yvette Cedex, France
Tel.: +33-1 69 08 79 25
Fax: +33-1 69 08 82 61
E-Mail: konrad.hinsen at cea.fr
---------------------------------------------------------------------





More information about the mmtk mailing list