Acceleration and Verification of Virtual High-throughput Multiconformer Docking

The work in this dissertation explores the use of massive computational power available through modern supercomputers as a virtual laboratory to aid drug discovery. As of November 2013, Tianhe-2, the fastest supercomputer in the world, has a theoretical performance peak of 54,902 TFlop/s or nearly 55 thousand trillion calculations per second. The Titan supercomputer located at Oak Ridge National Laboratory has 560,640 computing cores that can work in parallel to solve scientific problems. In order to harness this computational power to assist in drug discovery, tools are developed to aid in the preparation and analysis of high-throughput virtual docking screens, a tool to predict how and how well small molecules bind to disease associated proteins and potentially serve as a novel drug candidate. Methods and software for performing large screens are developed that run on high-performance computer systems. The future potential and benefits of using these tools to study polypharmacology and revolutionizing the pharmaceutical industry are also discussed