[gmx-users] com motion

[Christoph Freudenberger]

Mustafa ToprakXX wrote:
> Ok David,
> The temperature is what I want. According to my understanding the 
> update algorithm corrects the VCM and prevents accumulation of kinetic
> energy in the center of mass. When I checked the com xvg file this is
> clearly seen. Just like the temperature, the com koordinates does not 
> change much.
>   so it seems the code does not sample such kind of motion efficiently.
> Such a simulation may be sampled under external forces but that would
> be too unrealistic. Because it would be better to sample under intermolecular forces.

I don't really understand, what you accually want to do...
You're studing a ligand that is docked to some receptor protein and
want to see how much the ligand moves around the protein. is that
correct?
Are you doing this in vakuum or in a solvent? if do which one
and why is there no temp coupling for it in the mdp?

Concerning the COM motion: I gather from the mdp that you
keep the protein coords frozen. You have not defined a COMM group so
I suppose the COMM is computes for the whole system and (as the
mass of the protein is surely some orders of magnitude larger than
that of the ligand) the ligand motion is maybe just averaged out.
Did you check the protein ligand distance by MSD?

-- 
Christoph Freudenberger
Abt. Organische Chemie I AK Siehl - Uni Ulm -Tel: ++49-731-502-2785