[gmx-users] Restraining water

[Mark Abraham]

On 2/12/2010 12:51 PM, Guido Polles wrote:
> Hi,
> I know it looks a little bit strange, but i was trying to restrain
> just some water in my system.
> Now, if i put something like
> [ position_restraints ]
> ;  i funct       fcx        fcy        fcz
>     1    1       1000       1000       1000
> I put restraints on all the molecules.

Yes, that directive will pertain to all the molecules that have that [ 
moleculetype ]. I'm not sure if you only want position restraints on 
some of your water molecules.

> I thought about a turnaround,  just making a second water molecule
> type, but now complains about using more than one settle type.
> Is there any other way to do it?
> And if I follow the "Suggestion: change the least use settle
> constraints into 3 normal constraints" I have to worry about some kind
> of huge loss in performance?

I would expect the non-settle waters to use the solvent-optimized loops 
in the normal way, so there should be no difference in execution time 
between the unrestrained all-settle simulation and the partly-restrained 
party-settled simulation. Obviously, this is straightforward to test.

Mark