Christopher Lipinski - Analysis and hit filtering of a very large library of compounds screened against Mycobacterium tuberculosis.

Document created by Christopher Lipinski on Aug 22, 2014
Version 1Show Document
  • View in full screen mode

  Publication Details (including relevant citation   information): Ekins, Sean, Kaneko, Takushi, Lipinski,   Christopher A., Bradford, Justin, Dole, Krishna, Spektor, Anna,   Gregory, Kellan, Blondeau, David, Ernst, Sylvia, Yang, Jeremy,   Goncharoff, Nicko, Hohman, Moses M., Bunin, Barry A.,   Molecular bioSystems, 2010, 6  (11), pp 2316-2324

  Abstract: There is an urgent need for new drugs   against tuberculosis which annually claims 1.7-1.8 million lives.   One approach to identify potential leads is to screen in vitro   small molecules against Mycobacterium tuberculosis (Mtb). Until   recently there was no central repository to collect information   on compounds screened. Consequently, it has been difficult to   analyze molecular properties of compounds that inhibit the growth   of Mtb in vitro. We have collected data from publically available   sources on over 300 000 small molecules deposited in the   Collaborative Drug Discovery TB Database. A cheminformatics   analysis on these compounds indicates that inhibitors of the   growth of Mtb have statistically higher mean logP, rule of 5   alerts, while also having lower HBD count, atom count and lower   PSA (ChemAxon descriptors), compared to compounds that are   classed as inactive. Additionally, Bayesian models for selecting   Mtb active compounds were evaluated with over 100 000 compounds   and, they demonstrated 10 fold enrichment over random for the top   ranked 600 compounds. This represents a promising approach for   finding compounds active against Mtb in whole cells screened   under the same in vitro conditions. Various sets of Mtb hit   molecules were also examined by various filtering rules used   widely in the pharmaceutical industry to identify compounds with   potentially reactive moieties. We found differences between the   number of compounds flagged by these rules in Mtb datasets,   malaria hits, FDA approved drugs and antibiotics. Combining these   approaches may enable selection of compounds with increased   probability of inhibition of whole cell Mtb activity.

  Address (URL):