Computational Precision Oncology

We develop and apply computational approaches for analyzing and visualizing clinical, imaging, and omics data for (1) treating cancer using combination therapies adapted over time; and (2) discovering and understanding mechanisms of resistance and priming in cancer. This work integrates public cancer datasets with deep, longitudinal profiling of patient tumors to identify biomarkers that can aid in therapy selection and understand how cancers adapt to therapy. Current areas of focus include building multimodal single-cell tumor atlases, identifying spatial biomarkers in tumor ecosystems, and large-scale machine learning to understand mechanisms of therapeutic response and resistance in tumors.

Software Tools for Machine Learning in Cancer

We develop computational tools and infrastructure that can efficiently process large biomedical datasets and help investigators make sense of analysis results. We help lead the development of Galaxy, a Web-based scientific analysis platform that is used by thousands of scientists throughout the world to analyze large biomedical datasets, including genomic, proteomic, metabolomic, and imaging data. We are working to enhance Galaxy through visual analytics, machine learning, and cloud computing. We are also enhancing large-scale data commons through our work to develop the NHGRI Genomic Data Science Analysis, Visualization, and Informatics Lab-Space (AnVIL).