MULTIVARIATE ANALYSIS FOR MOLECULAR TOXICOLOGY SCREENING

DESCRIPTION (provided by applicant): Automation of drug screening assays has greatly increased the speed and identification of drug "hits" with potential therapeutic value. Pharmaceutical companies are finding it very costly to test the thousands of "hits" for safety in animals as a prelude to human trials. The lack of a quick, reliable and efficient in vitro test for drug toxicity before a compound is submitted to animal and human testing costs the pharmaceutical industry billions of dollars each year. VistaGen has proprietary expertise in applying in vitro stem cell assays to predict drug toxicity at an early stage in preclinical drug development. The company is developing a simple, sensitive, and reproducible screening technology for in vitro toxicity assessment that generates molecular fingerprints induced by toxic drugs as assessed by selective changes in genomic and proteomic expression patterns in stem cells measured using cDNA microarrays and a protein SELDI-TOF microchip technology. At the heart of the predictive capabilities of this technology is a novel biomathematical system which will provide the data analysis needed to interpret the molecular profiles derived from the in vitro screening assays and associate them with known toxicological "fingerprints." An integrated informatics superstructure will allow the data to be linked to in vivo drug toxicity, human clinical, and chemical databases for analysis of potential structure activity relationships. The objective of this application is to develop this biomathematical component. Proteomic and RNA data derived from ongoing stem cell drug screening studies will be used to develop the mathematical and statistical analysis systems that will test the existence and reproducibility of patterns and subpatterns related to possible toxicology of the compounds tested. Validation of our novel analysis systems will be tested using standard cross-validation techniques and Monte Carlo simulations. The primary outcome of these initial studies will be a series of data analysis tools and algorithms that will be applied to more complete data sets in the future. The finalized methodologies, including a completed informatics infrastructure will become part of VistaGen's commercial product. PROPOSED COMMERCIAL APPLICATION: Our technology will allow pharmaceutical companies to prioritize and focus on those in vitro drug "hits" that are most likely to be efficacious and safe in vivo in animals and humans. Since drug development failures can cost the pharmaceutical industry billions of dollars, we anticipate, based on direct discussions with the industry, that through licensing and partnering with pharmaceutical companies, that the value of the commercialization of our final GenesisScreen system will be worth millions of dollars. PROPOSED COMMERCIAL APPLICATION: