Computer-aided approach for search of phytomedicines with the required properties

Many pharmaceuticals developed on the basis of synthetic chemistry exhibit adverse effects and toxicity, which restricts their use in medical practice. Drug discovery from medicinal plants are often considered as a prospective leads for novel pharmaceuticals due to their mild biological action and less side effects and toxicity. Our computer program PASS (Prediction of Activity Spectra for Substances) predicts pharmacological activity profile on the basis of chemical structure (J. Comput. Aid. Molec. Des., 2002; 16:819; http://ibmc.msk.ru/PASS). PASS provides reasonable accuracy of predictions that was shown for both synthetic compounds (J. Med. Chem. 2003; 46:3326) and natural products (Mini-Rev. Med. Chem. 2005; 5:319). Based on computer prediction, it is possible to identify the substances with the required pharmacological activity profile among the compounds presented in databases of natural products. It automatically provides a link to the medicinal plant that contains a particular compound, leading to discovery of new phytomedicines. Jointly with specialists from JSC “Diod“ (http://www.diod.ru) based on computer prediction new pharmacological properties of plant flavonoid dihydroquercetin and other substances were discovered, that were confirmed in parallel randomized placebo-controlled clinical trials.