A White-Box Machine Learning Approach for Revealing Pathway Mechanisms


Event details

Date and time 22.01.2020 11:0012:00  
Place and room
Speaker Prof. Jason H. Yang, Rutgers New Jersey Medical School, Newark, NJ (USA)
Category Conferences - Seminars
Recent advances in high-throughput experimental technologies and data analyses now enable unprecedented observation, quantification and association of biological signals with cellular phenotypes. However, current approaches for interpreting large biomedical datasets are unable to provide casual, mechanistic biological understanding. Here, we will describe a “white-box” machine learning approach integrating prospective cellular network modeling with machine learning to identify experimentally testable pathway mechanisms from biochemical screening data.

We will demonstrate how this approach enabled the novel discovery that purine biosynthesis is involved in bactericidal antibiotic lethality, through its coupling to central carbon metabolism. We will discuss how such approaches may be extended towards advancing systems medicine by revealing mechanisms underlying disease pathogenesis and therapeutic efficacy. We propose such approaches may be generalized to investigate any quantifiable cellular phenotype using relevant biological networks.

Dr. Jason Yang is a new Assistant Professor and Chancellor’s Scholar in the Department of Microbiology, Biochemistry and Molecular Genetics and in the Center for Emerging and Re-Emerging Pathogens at Rutgers New Jersey Medical School. He received his Ph.D. in Biomedical Engineering from the University of Virginia, where he trained with Dr. Jeffrey Saucerman, studying β-adrenergic signaling in cardiac myocytes. He completed his postdoctoral training with Dr. James Collins at MIT and the Broad Institute, studying metabolic mechanisms of antibiotic-induced bacterial death physiology. Jason leads a systems biology research group, where they are developing approaches that integrate high-throughput experimentation with network modeling and machine learning to accelerate the discovery of causal biological mechanisms as they pertain to the pathogenesis and treatment of chronic and infectious diseases.