The second pilot in the Department of Energy (DOE) and National Cancer Institute (NCI) joint-pilot program—RAS Structure and Dynamics in Cellular Membranes—aims to develop a targeted therapy for RAS-driven cancer initiation and growth using innovative molecular dynamics simulations at unprecedented scales in an unsupervised machine-learning ecosystem that analyzes simulation data, creates new hypotheses, and returns new results to higher-scale simulations. RAS gene mutations are the leading cause of cancer in about a third of cancer cases. With the integration of multiscale, physics-based dynamics simulations, data analytics, machine learning, hypothesis generation, and advanced computing architectures, Pilot 2 will help to improve health and treat disease by enabling a fundamental understanding of RAS activation and predicting the success of new oncology therapies.
The pilot supports NCI’s mission by increasing the understanding of cancer biology and allowing scientists to identify possible targets for prevention. It also supports the DOE’s mission by developing new multiscale molecular dynamics methods and codes for systems used in the Collaboration of Oak Ridge, Argonne, and Lawrence Livermore (CORAL). In addition, it will provide scalable tools for integrating experimental data with simulations from ultrahigh performance supercomputers.