MEDI John Macor  Wednesday, March 19, 2014 

229 - Prediction of activity spectra of substances (PASS): Twenty years of development

Vladimir Poroikov, vladimir.poroikov@ibmc.msk.ru, Dmitry Filimonov, Alexey Lagunin, Tatyana Gloriozova. Department for Bioinformatics, Orekhovich Institute of Biomedical Chemistry of Rus. Acad. Med. Sci., Moscow, Moscow Region 119121, Russian Federation

Among the numerous ligand-based drug discovery tools PASS occupies a special place because its development has been started over 20 years ago (Poroikov et al. Automatic Documentation and Mathematical Linguistics, 1993, 27: 40-43). During the past years PASS is improved and extended permanently (Filimonov and Poroikov, In: Chemoinformatics Approaches to Virtual Screening, RSC Publishing, 2008, 182-216). Current PASS version predicts 6,400 biological activities with a mean accuracy of about 95% based on structure-activity relationships elucidated from the training set consisted of 330,000 biologically active compounds. Freely available online resource (http://www.way2drug.com/passonline) is used by ~9,000 researchers from ~90 countries. Over 300,000 predictions were performed; over 50 independent published studies confirm PASS predictions by subsequent synthesis and biological testing.

Using PASS predictions, novel pharmaceutical agents have been discovered with anxiolytic, anti-inflammatory, antihypertensive and other actions. By PASS application to the antihypertensive drugs Perindopril, Quinapril and Monopril we identified a nootropic action, which is likely not related to their antihypertensive effect. To find new anticancer agents, we have analyzed dozens of millions of structures from ChemNavigator database and selected a few dozen compounds for biological testing. Two out of eleven tested compounds were found to be potent anticancer NCEs, with synergistic action to the known p53 reactivator RITA. PASS application significantly increases the chances for discovery of new more safety and potent pharmaceutical agents, and predicts biological activity profiles of drug-like substances in chemical biology.

Acknowledgement. This work was partially supported by FP6 grant No. LSHB-CT-2007-037590, RFBR grants No. 12-04-91445-NIH_a/RUB1-31081-MO-12, 12-07-00597_а and 13-04-91455-NIH_a.


Wednesday, March 19, 2014 02:50 PM
General Oral Session (01:30 PM - 05:10 PM)
Location: Dallas Convention Center
Room: BLRM C1

 

*ACS does not own copyrights to the individual abstracts. For permission, please contact the author(s) of the abstract.

 

Close Window