CCL: SAR predictions for this paticular question

 Sent to CCL by: "Derek J. Cashman" [derek.cashman%gmail.com]
 On Fri, Aug 8, 2008 at 4:12 PM, Nancy Ann Neale nealen!A!mail.nih.gov
 <owner-chemistry+*+ccl.net> wrote:
 >
 > Sent to CCL by: "Nancy Ann Neale" [nealen#,#mail.nih.gov]
 > I am looking for recommendations for SAR softawre for a specific problem.
 >
 > What we have at this point is approximately 50 compounds that have bee
 qualitately analyzed for activity in a certain cell.  We haven't structural data
 on the surface protein that shuttles substrates accross the membrane, but that
 is only the first step in detrmining what a good canddiate is.
 >
 > We have 'downstream' data on these compounds with respect to activity,
 IC50, and a selectivity ratio.
 >
 > My first thought is AutoDock, but am concerned that may not do the trick
 because of our lack of structural data of that protein.
 >
 > My goal is to train some program with the good amount of data we have to
 recognize adn ultimately predict reasonable candidates in the future for this
 application.
 >
 > Any suggestions are most welcome.
 Looks like this data set would be ideal for running a CoMFA study
 (Comparative Molecular Field Analysis) in SYBYL (www.tripos.com).
 Assuming that SYBYL is available or within your budget (it's a
 commercial package).
 Derek J. Cashman, Ph.D.
 derek.cashman+*+gmail.com
 "A Drug is any substance which, when injected into a rat, produces a
 publishable, scientific paper."