[gmx-users] Coupling groups

[Mark Abraham]

On 6/03/2012 8:21 PM, Steven Neumann wrote:
> Dear Gmx Users,
> I am running my protein with ligand. at the begining of my simulation 
> the protein is far away from the ligand. Then in about 5-10 ns my 
> ligand binds to the protein surface remaining there for rest 90-95 ns. 
> Would you adjust coupling groups in thermostat as:
> a) Protein_LIg and Water_and_ions
> b) Protein Water_ions_LIG
> c) Protein LIG Water_ions
>

I'd consider a single coupling group before any of those. See final part 
of manual 3.4.8 and ref therein. c) is unsound because the ligand is 
likely too small. The lack of T-group atom contiguity in a) might be an 
issue, I'm not sure. b) is probably what people do, but you know this 
from your literature searching about similar work, right? :-)

Mark