Open Competition 3 - Information Technology
New Software Tool for Improving Drug Discovery and Development
Develop new simulation software that uses quantum mechanics to evaluate molecular forces and electronic structures in organic and biological systems to help improve the development of drugs by accurately predicting biological and toxicological effects.
Sponsor: Targacept Inc.200 East First Street
Winston Salem, NC 27101-4165
Sophisticated simulation software could transform the current costly and significantly time consuming process of new drug discovery that relies heavily on prototyping as well as trial and error. The average time to bring a new drug to market is 12 years, and two thirds of the approximately $800 million cost is spent on drugs that ultimately fail to meet necessary efficacy and safety levels. Reducing this failure rate through identification of potential problems during the early stages of drug design is far more cost-effective than realizing failure during clinical trials. In pursuit of this goal, Targacept will bring highly accurate simulation capability to the field of computational drug design through the implementation of Car-Parrinello molecular dynamics, also known as ab initio (from the beginning) molecular dynamics (aiMD) within a full simulation software system. The aiMD method describes the dynamical properties of matter with great accuracy because it allows the simulation of molecular behavior and chemical systems based on their quantum mechanical (QM) properties. This new software tool suite will facilitate the identification and creation of viable molecular candidates that have desired drug-like physico-chemical and toxicological properties through the use of QM calculations conducted in dynamic, physiologically relevant conditions. This new tool for computational drug design could be as revolutionizing within the pharmaceutical industry as computer-aided design and computer simulation were to aerospace and automotive industries. The Princeton Materials Institute at Princeton University (Princeton, NJ) and the Department of Chemistry at the University of Colorado (Denver, CO) will collaborate on this three-year project. Extensive evidence supports the predictive power of aiMD in materials and protein science, but aiMD is unproven in assessing drug candidates. Developing and implementing tools for conformational search and analysis as well as techniques for reaching computational speed and accuracy necessary for usability are among the high technical risks faced by the team. ATP funding will greatly accelerate the development and delivery of this software system s potential benefits. Successfully developed, aiMD software will provide better, early predictions about the molecular properties of drug candidates, improving selection of compounds and reducing failures later in the drug development process. This in turn could reduce the cost of bringing a new drug to market by an estimated 25 percent, or $200 million, and cut the time to market by two to three years. If aiMD saves the pharmaceutical industry $2 billion annually in R&D as estimated, then patients should benefit from consequent reductions in prescription costs for new drugs. Though harder to quantify, improvements in health and quality of life - such as more effective therapies, fewer lost work-days, and shorter hospital stays - could also result. Lastly, the software will likely find wide applicability in the fields of chemistry, structural biology, molecular biophysics, molecular recognition and enzymology.