Romain Lopez
Affiliation
New York University
Talk Title
Causal Networks and Applications to Molecular Biology
Abstract
While causal graph discovery seems appealing for biological data analysis, these approaches have had limited impact on single-cell and molecular data analysis. We examine three key barriers: restrictive assumptions in causal models, scalability limitations, and identifiability challenges in real biological datasets. In this overview, we discuss these obstacles, highlighting recent progress as well as emerging opportunities.
Bio
Romain Lopez is an Assistant Professor with a joint appointment in Computer Science at the Courant Institute and in Biology at New York University. His research focuses on the intersection of statistics, computation, and modeling with applications to biological systems. Romain leads the NYU Biological Machine Learning Group, developing statistical methods and machine learning tools for analyzing single-cell omics data. He is known for creating single-cell variational inference (scVI), a scalable framework for probabilistic representation and analysis of single cell genomics data. Before joining NYU, Romain completed his postdoctoral training jointly at Stanford University and Genentech Research & Early Development, working with Jonathan Pritchard and Aviv Regev. He received his PhD in Computer Science from UC Berkeley, where he was advised by Mike Jordan and Nir Yosef. His work aims to enhance our understanding of cellular processes and transform drug discovery through computational approaches that address causality, interpretability, and complex interactions in biological systems.