AMBER Archive (2007)

Subject: AMBER: Rdf question

From: BJÖRN KARLSSON (BJORN.KARLSSON_at_hik.se)
Date: Sun Jun 10 2007 - 14:51:16 CDT

Hi!
I am currently simulating (at NPT) a mixture of multiple copies of an organic molecule (10 copies) which is a drug and acetic acid (100 copies) in an organic environment consisting of approximately 1000 molecules of explicit toluene using periodic boundary condition at 5ns.

My main interest from this simulation is to be able to extract a rough estimate of the optimal potential interaction energy and the average number of e.g. acetic acid molecules that are around each drug molecule during the simulation.
I now that radial distribution functions is a good way to extract this kind of information however when I've tried to obtain g(r) values for the probabilities of finding a specified atom of acetic acid around another atom of the drug (plotting g(r) for all solutes at the same time i.e. solute=drug no. 1-10 and solvent=acetic acid no. 11-111) the values I obtain in the standard-rdf-file using the user-input density value of pure acetic acid instead of default water density my rdf doesn't approximate 1 for high values for large atom distances. What am I doing wrong?
Once I have obtained a nice looking rdf is it then possible to get the rough estimate of the optimal interaction distances and energies by using the potential of mean force (PMF) by using the expression PMF=-kTg(r) and recalculate the rdf to a PMF plot?

Cheers/Björn C. G. Karlsson

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu