SBIR Phase I: The Development of Self-Organizing Maps for Drug Discovery

This Small Business Innovation Research Phase I project will test the feasibility of using Self-Organizing Maps (SOMs), a non-linear, topology-preserving pattern recognition technique, in the design of small molecules with biological activity. Due to a serious unmet need for computer-based tools that can accurately predict early development issues such as potency, safety, and efficacy of drug candidates, the pharmaceutical industry currently relies almost exclusively upon expensive prototyping to evaluate candidates. Many drug candidates fail only after costly pharmaceutics development and/or clinical trials, limiting the ability of drug companies to address anything beyond the most profitable markets. The early stage technology proposed in this Phase I project will significantly advance the state-of-the-art of drug design by addressing one of the major problems with which current computer-aided drug design has long struggled: how to model the non-linear feature-target relationships with dependent features commonly associated with biological systems. The commercial application of this project is in the area of drug discovery and development. The innovative software to be developed through this project will allow the drug companies to computationally evaluate a vast number of drug candidates, resulting in more candidates being investigated and more effective drugs discovered.