In Silico Screening of Chinese Herbs Using Random Forest

Filed in archive Drugs, Vaccines and Therapeutics , Genomics, Proteomics and Bioinformatics on January 25, 2007

Researchers from the King's College London have conducted the first large-scale computer screenings of chinese herbs, which revealed a wide variety of compounds with potential for use in treating HIV/AIDS, cancer, Alzheimer's Disease, arthritis and other diseases. They used Random Forest, a form of multiple decision trees, has been used to screen a database of Chinese herbal constituents for potential inhibitors against several therapeutically important molecular targets.

About 62 percent of the species were found to contain chemicals with characteristics required for activity against at least one disease and 53 percent against two or more diseases. The authors issued a caveat, however, that further in vivo experiments will have to be performed to confirm their findings.

More focused studies, using techniques such as ligand-receptor docking, are also required, as are theoretical approaches to questions such as the possible role of weak inhibitors in modulating metabolic or signal transduction pathways, something which impacts directly on the use of many herbs and may furthermore provide new insights into the development of next-generation pharmaceuticals without some of the problems currently associated with strong, selective inhibitors.

Predictions are made here largely on the basis of in vitro experiments. Further work, particularly on the absorption, distribution, metabolism, and excretion of plant compounds, is needed to assess the role such predictions may have in vivo.

For further info, refer to the study,Virtual Screening of Chinese Herbs with Random Forest, published at the Journal of Chemical Information and Modeling.