This Small Business Innovation Research (SBIR) Phase I project will demonstrate that chemical compound activity/binding prediction can be significantly improved by applying geometric mining to the Joint Pharmacophore Space (JPS). Activity/binding prediction is of immense importance to the pharmaceutical industry as it allows evaluating biological properties early on in the drug development process. The proposed technique applies novel 3D clustering and statistical analysis to model configurations contributing/preventing binding between potentially large numbers of compounds and pharmacophoric targets (the JPS) and uses these models to search for other compounds with similar activity. In addition, the technique allows the interactive study of "binding clusters" in order to design new compounds with specific binding properties and physicochemical properties. The broader/commercial impacts of this research are (1) improved efficiency of drug discovery pipelines, opening new opportunities, accelerating time-to-market of new drugs, and cutting R&D costs for pharmaceutical and biotech companies, and (2) the furthering of the understanding of basic compound properties, such as blood-brain-barrier permeability. This could have far-reaching consequences in the fight against diseases. Yet another outcome of the successful completion of this project is (3) the specific identification of new lead compounds antagonizing the toxic compounds causing Alzheimer's. Consequently, the project has a significant market potential as well as important ramifications for the improvement of human health.
