BindingDB: Data and Tools for Drug Discovery, Chemical Biology and Systems Pharmacology

点击翻译按钮获取中文摘要

Abstract Most medications are small organic molecules which work by binding disease-related proteins. As biomedical research yields ideas for additional proteins which could be targeted by new drugs, thousands of small molecules are synthesized and tested, in both industry and academia, for their ability to bind these targeted proteins. Chemical biologists also develop small molecules targeted to specific proteins in order to elucidate pathways and cellular mechanisms. The resulting flow of protein-ligand interaction data have many uses well beyond the specific projects which generated them, such as testing and improve computational drug design technologies, and generating hypotheses for the mechanisms of action of bioactive compounds. However, they are published almost exclusively in scientific articles and patents, where they cannot be readily and systematically searched, browsed, and retrieved. BindingDB, the first open database to collect these protein- small molecule binding data and make them available for query, download, and analysis, provides users worldwide with ready access to a growing data collection, currently 1.2 million data for 500K small molecules and thousands of proteins. In the forthcoming grant period, we plan to continue the core activity of expanding BindingDB’s data collection by harvesting relevant data from other on-line resources and continuing in-house curation of data from new scientific articles and patents. We will also carry out major upgrades of the BindingDB web-site, providing a less cluttered look and new features, such as ones to refine search results, gain overviews of the large BindingDB data collection, and customize the content of data downloads requested by users. New capabilities will also be developed to support applications in the emerging field of systems pharmacology, by annotating the proteins in biomolecular networks and pathways with small molecule binding data from BindingDB and constructing tools to qualitatively model the effects of perturbing pathways with small molecule binders. Finally, we will share information about BindingDB through a variety of channels, and will work with education experts to develop and disseminate instructional applications of BindingDB.