Learning objectives

• Cellular heterogeneity in tumour microenvironments provides crucial information for understanding disease mechanisms and predicting progressions
• A probabilistic topic model that can identify the underlying cellular states and their interactions from single-cell RNA-seq data
• To achieve necessary scalability in time and memory, we developed a simple approximation step that can create manageable pseudo bulk data from single-cell data
• Applying the same principle, we can dissect millions of cell-cell interactions to summarize them into interpretable key cellular interaction topics

Speaker biography

Dr. Yongjin Park is an Assistant Professor at the University of British Columbia and BC Cancer Research. He holds Canada Research Chair Tier-2 (Integrative Causality Inference of Cancer Mechanisms). He received his Ph.D. in Biomedical Engineering at the Johns Hopkins University and postdoctoral training in Computer Science and Artificial Intelligence Laboratory at Massachusetts Institute of Technology. His research focuses on applying causal inference methods in large-scale single-cell genomics data to understand causal mechanisms of cancer development and progression.