Building the Mind of Cancer
Trey Ideker
September 7, 2022, Wednesday, 3:00 PM - 4:00 PM EST
Abstract
Most drugs entering clinical trials fail, often related to an incomplete understanding of the mechanisms governing drug response. Machine learning techniques hold immense promise for better drug response predictions, but most have not reached clinical practice due to their lack of interpretability and their focus on monotherapies. To address these challenges, I will describe development of interpretable deep learning models of human cancer cells. These models are trained on the responses of thousands of tumor cell lines to thousands of approved or exploratory therapeutic agents. The structure of the model is built from a knowledgebase of molecular pathways important for cancer, which can be drawn from literature or formulated directly from integration of data from genomics, proteomics and imaging. Based on this structure, alterations to the tumor genome induce states on specific pathways, which combine with drug structure to yield a predicted response to therapy. The key pathways in capturing a drug response lead directly to design of synergistic drug combinations, which we validate systematically by combinatorial CRISPR, drug-drug screening in vitro, and patient-derived xenografts. We also explore a recently developed technique, few-shot machine learning, for training versatile neural network models in cell lines that can be tuned to new contexts using few additional samples. The models quickly adapt when switching among different tissue types and in moving to clinical contexts, including patient-derived xenografts and clinical samples. These results begin to outline a blueprint for constructing interpretable AI systems for predictive medicine.
Bio
Trey Ideker, PhD, is a UCSD Professor of Medicine, Bioengineering and Computer Science, and former Chief of Genetics. He is a Co-Director of the Bioinformatics and Systems Biology PhD Program, UCSD. He directs or co-directs the National Resource for Network Biology, and the Cancer Cell Map and Psychiatric Cell Map Initiatives. Trey received BS and MEng degrees in Computer Science from MIT and his PhD in Genome Sciences from the University of Washington under Drs. Lee Hood and Dick Karp. Trey is a pioneer in genomic, transcriptomic, and proteomic analysis and in the theory and practice of Systems Biology. He founded and continues to develop the widely used Cytoscape network analysis platform (>30,000 citations). His lab also created the Hannum epigenetic clock, the first to measure human aging rates using DNA methylation. Trey is on the Board of Scientific Advisors to the National Cancer Institute and formerly the National Human Genome Research Institute. He serves on the editorial boards of Cell, Cell Systems and PLoS Computational Biology. Trey was named a Top 10 Innovator by Technology Review, received the 2009 ICSB Overton Prize, and is an AAAS & AIMBE Fellow. He is a Web of Science Highly Cited Researcher (top 1% by citations).