Seminar: Predictive, explainable, and autonomous AI for single-cell multi-modality biology

Abstract

Current biotechnologies can simultaneously measure multiple modalities (e.g., gene expression and electrophysiology) from the same cells. AI holds great promise for fully understanding such data, inferring how genes regulate cellular diversity, function, and disorder. This talk will cover my work and vision of how predictive, explainable, and autonomous AI models can enable data-driven single-cell biological insights, including navigating hypotheses for gene-to-function mapping and in silico perturbations of cell behavior that closely mirror the wet lab experiment. Finally, I will expand the definition of multi-modality and present my roadmap for building cellular digital twins.

_Biography

_{Xin Tang is from the Broad Institute of Harvard and MIT and the Department of Bioengineering at Harvard School of Engineering and Applied Sciences, working with Prof. Jia Liu and Xiao Wang. His interests include developing and applying state-of-the-art tools in artificial intelligence and single-cell biology to understand how individual cells work in their cell-type-specific, multi-modal, and spatiotemporal context. He also has a special interest in brain-computer interfaces. His work has been published in venues including Nature, Cell, Nature Communications, Nature Electronics, Nature Methods, and Nature Neuroscience.}