[gmx-users] frozen atoms

[David]

On Fri, 2002-06-21 at 21:11, Valentin Gogonea wrote:
> Dear gromacs users,
> 
> In the manual is written that the positions of frozen atoms are not 
> updated. Does this mean that gromacs still assigns velocities to frozen 
> atoms, calculate new positions but ignore them? Does this happen when 
> the atoms are included into an energy exclusion group?
> 
> Does it make sense to couple to the thermal bath a frozen group? If yes 
> what will be the coupling temperature? 0K or 300K?
To get really frozen atoms you have to make a center-of-mass motion
removal group as well for the same atoms. If your freeze atoms are part
of a larger group that undergoes center of mass motion removal, then
your frozen atoms will move too. There is no problem incluing frozen
atoms in shake etc. If the group is large you may want to define an
energy exclusion group, to avoid computation of the interaction within
the frozen group. A drawback is that you will not be able to compare
energies to a non-frozen simulation.

This is poorly documented I'm afraid. In principle it could be automated
(make the freeze group automatically a com group), but I am not entirely
sure whether that is a good idea (because I didn't think it through
yet).

-- 
Groeten, David.
________________________________________________________________________
Dr. David van der Spoel, 	Biomedical center, Dept. of Biochemistry
Husargatan 3, Box 576,  	75123 Uppsala, Sweden
phone:	46 18 471 4205		fax: 46 18 511 755
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://zorn.bmc.uu.se/~spoel
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++