DescriptionThe Cancer Moonshot was established in 2016 with the goal of doubling the rate of progress in cancer research. A major component is the strategy to use modeling, simulation, and machine learning to advance our understanding of cancer biology and to integrate what we know into predictive models that can guide research and therapeutic developments. In 2015, the U.S. Department of Energy (DOE) formed a partnership with the National Cancer Institute (NCI) to jointly develop advanced computing solutions for cancer by bringing together researchers from four DOE laboratories (Argonne, Los Alamos, Livermore, and Oak Ridge) with the Frederick National Laboratory for Cancer Research (FNLCR). This integrated team has launched three pilot projects, each addressing a major challenge problem on the forefront of precision oncology: (1) provide better understanding and eventually develop new drugs for the RAS oncogene family of cancers which impact 30% of cancers; (2) develop models that can predict tumor response to drugs to enable physicians to more precisely target an individual patient’s tumor; and (3) analyze electronic medical records of millions of cancer patients to streamline the introduction of new precision oncology therapies. The CANDLE (CANcer Distributed Learning Environment) project aims to develop an end-to-end computational environment to bring deep learning to these key problems. This talk introduces the cancer moonshot, describes the three overarching cancer problems, and provides a roadmap of the CANDLE project. We will discuss the impact of deep learning, advanced modeling, and simulation on cancer research and future approaches to treating patients.