Automated, Web Based Software Development for Predictive QSAR Modeling

DESCRIPTION (provided by applicant): In the course of previous basic studies in our laboratory, we have developed statistically significant and robust Quantitative Structure-Activity Relationships (QSAR) methodologies, which incorporate rigorous validation procedures and lead to models with a high predictive power and practical utility. The methodologies are built upon the similarity principle, i.e., similarity or diversity of chemical structures determines similarity or diversity of their biological action. Thus, our methodologies employ variable selection procedures aimed at identifying descriptors most relevant with respect to the target property. Paramount to our approach is rigorous model validation with external datasets which ensures the highest hit rates when predictive QSAR models are ultimately applied to screening chemical databases or virtual libraries for biologically active compounds. Pilot studies have indicated that our approaches to QSAR modeling can be organized and implemented in a functional workflow using specialized middleware giving rise to a potentially important commercial QSAR technology The main objective of this STTR project is to transfer the prototypical technology built in the Laboratory for Molecular Modeling at UNC-Chapel Hill to Phorcast, Inc. The ultimate objective of this project is to build and deploy an advanced, autmoated web-based QSAR server, which is capable of utilizing vast grid technologies such as those in the North Carolina Grid project. The server will allows researchers to easily and efficiently build predictive QSAR models for experimental drug discovery projects. Then end goal is to develop a commercial product that automates QSAR model generation, stores and provide access to validated pharmacological, toxicological, and other property (e.g., ADME) predictors, and affords virtual screening of public and proprietary chemical databases to identify higly significant computational hits. The key components to design the server are addressed in the following Phase 1 Specific Aims: 1. Develop and enhance an integrated QSAR workflow that incorporates QSAR model building algorithms with stringent model validation protocols for application to database screening. 2. Incorporate a QSAR web server into the NC Grid or a UNC cluster. 3. Incorporate the latest advancements in QSAR and database screening technology to meet the needs of the research community.