On Wed, Jan 09, 2013, Алексей Раевский wrote:
> I want to create a system with hybrid solvent (explicit/implicit). My
> system is large enough to calculate it with explicit SOL. But it is
> critical to study the behavior of several water molecules in the site. The
> idea is to generate a layer of explicit water molecules around the protein
> molecule with ligand in the binding site and than fill the box with
> continuous imlicit solvent with epsilon = 80. It will reduce a
> computational time. To prevent the departure of discrete water molecules
> into the implicit layer I want to limit the distance of their movement with
> something like distance constraints or anything else. Is it real or I have
> to find another way or choose another software for this purpose.
Amber doesn't implement this sort of calculation, although some other
simulation programs have things somewhat like this. (Most often, the
implicit solvent desciption goes out to infinity, rather than to the edges
of periodic box.) Depending on the size of your system, you may well find
that a hybrid scheme like this does not, in fact, reduce computation time,
but that is something that depends on all kinds of variables, such as what
computer, what program, and size of your system.
...dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Jan 09 2013 - 06:30:03 PST
