[gmx-users] Saving CPU time with constraints ?

[David]

On Sat, 2003-03-15 at 15:57, Ing. Vojtěch Spiwok wrote:
> Dear gromacs users/developers
> 
> I would like to run an MD of ligand bound in the binding site
> of a protein with most atoms of the protein constrained (frozen).
> Does GROwingMonstersAndCloningShrimps calculate an energy
> for all atoms in every step even if they are constrained ? In other
> words, does it worth to delete those atoms which are frozen and
> too distant to site of interest to save CPU time ?
> I appologise if this information is easy to find in manual or on
> program site, I didn't succeed.

You can define energy_grp_excl = frozen
in the mdp file (check exact syntax) in which case no nonbonded
interactions will be computed for those groups internally. However
interactions of the frozen group with rest of the system is computed. If
you delete those atoms, you miss out on the forces they yield on your
part of interest. If you freeze half of the system, you can hence reduce
the number of interactions by one quarter.


-- 
Groeten, David.
________________________________________________________________________
Dr. David van der Spoel, 	Dept. of Cell and Molecular Biology
Husargatan 3, Box 596,  	75124 Uppsala, Sweden
phone:	46 18 471 4205		fax: 46 18 511 755
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://xray.bmc.uu.se/~spoel
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++