On Jun 26, 2014, at 2:06 PM, Brian Radak <radak004.umn.edu> wrote:
> I'm looking to do simple comparisons of 3D-RISM and NLPB calculations of a
> large biomolecule in solvent. Although solute-solvent pair distribution
> functions are natural outputs of both types of calculations, rism3d.snglpnt
> does not seem to readily output the total electrostatic potential (although
> presumably it *must* calculate this at some point, even if it includes
> contributions from Lennard-Jones terms).
>
3D-RISM does not, in fact, calculate the electrostatic potential on a grid. It does calculate the electrostatic potential energy at each grid point between the solute and each solvent site (O, H, Na+, Cl-, etc.). This is output with the LJ potential energy using the -uuv flag. However, I don’t think this is what you want.
> I suppose my question is this: how can I calculate the electrostatic
> potential on the 3D-RISM grid? Presumably there is some relation (although
> perhaps not at all trivial) to the charge density (--quv) and/or charge
> distribution (--chgdist), but the manual does not seem to indicate what
> these quantities actually are (are they related to the solvent site PDFs?
> do they include solute contributions in seem charge smearing manner?) How
> are the charge density and charge distribution different? If they are
> trivially different by a factor of one over the cell size, that seems like
> a somewhat frivolous output…
>
If you want an electrostatic potential comparable to that from PB, you will need to calculate it yourself. Basically, for each point on the grid, you need to integrate the charge density over all space divided by the separation. This includes both the solvent contribution (—quv) and the solute. The relevant equation in my second edition Griffiths electrodynamics is 2.24.
Hope this helps,
Tyler
-----------------------------------
Tyler Luchko
Assistant Professor
Dept. of Physics & Astronomy
Cal State Northridge
Northridge, CA
tluchko.csun.edu
818-677-6442
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Jun 27 2014 - 12:00:02 PDT
