Leading Edge Predictors for Drug Discovery

CSLogWS Home

CSLogD Home

CSLogP Home



CSGenoTox Home



     ...Dataset Profile

Dataset Used in Generating the CSPB Model

Sources for Experimental Protein Binding Values:

The activity values used in the generation of this model were taken from the following sources.  Agreement of values reported for the same compound was sufficient to justify combining the data from these sources.

    The Pharmacological Basis of Therapeutics, 9th Edition, appendix

    AHFS Drug Information

    Nurse's Drug Guide

    Allgemeine Pharmakologie und Toxikologie

CSPB Compound Profile

Drug like / non-drug like designations:

The compound profile listed below identifies only 86% of the compounds used for the development of this predictor as "drug-like", though virtually all are commercially available drugs.  The reason for this discrepancy is that some drugs are non-oral, and based on Lipinski's "rule of fives" for permeability and solubility of oral therapeutic agents, some non-oral drugs will not be designated as drug-like. 

(see CSpKa Dataset Profile for an explination of the, "Rule of Fives")

The predominant structural feature in the training set is the aromatic ring system, followed by unsaturated N-heterocyclic systems. Both types of ring systems are associated with fused ring formations in the majority of compounds.

A table of compound heterogeneity and plots of the major groups and ring types are given below.

Back to: CSPB  Home Page

user login
contact us

To contact us:

Phone: 978-501-0633

Fax: 781-275-5197

Email:  sales@chemsilico.com

Copyright © 2003 ChemSilico LLC All Rights Reserved

Terms and Conditions of Use | Privacy Policy

ChemSilico is a registered trademark of ChemSilico LLC, Tewksbury, MA 01876