报告题目：Model-based deep embedding for the analysis of single-cell RNA sequencing data

报告人：韦智  教授（新泽西理工）

主持人：周爱民  教授

报告时间：2023年11月17日（星期五）13:00-14:00

报告地点：必赢BWIN普陀校区孟刘馆二楼报告厅

报告摘要：

Methods to sequence the DNA and RNA of single cells are poised to transform many areas of biology and medicine. Single-cell RNA sequencing (scRNA-seq) promises to provide high resolution of cellular differences. Clustering transcriptomes profiled by scRNA-seq has been routinely conducted to reveal cell heterogeneity and diversity, followed by differential expression (DE) analysis to identify marker genes accounting for cellular differences. However, the analysis of scRNA-seq data remains a statistical and computational challenge, due to the pervasive dropout events, which obscure the high-dimensional data matrix with many 'false' zero count observations. Most existing methods focus on dimension reduction, often followed by simple clustering using, for example, k-means. Such a divided strategy is suboptimal for clustering, as we demonstrate. Furthermore, subsequent differential expression analysis after clustering incurs the so-called “double use of data problem, which will compromise type 1 error control for standard statistical tests. In this talk, I will introduce model-based deep autoencoders to address these issues. The proposed approaches leverage the most recent developments in feature representation learning in deep learning and feature selection in statistical learning, as well as prior information from domain scientists. This development provides a good example of how computer science, statistics, and domain science integrate to yield the optimal solution. Extensive experiments on both simulated and real datasets demonstrate that the proposed methods can boost clustering performance significantly while effectively filtering out most irrelevant genes. Our methods can generate more biologically meaningful clusters, enhancing interpretability as desired by biologists.

报告人简介：

Dr. Wei is a Professor of Computer Science and Statistics (joint appointment) at the New Jersey Institute of Technology. He is also an adjunct Professor at the University of Pennsylvania. Dr. Wei received his Ph.D. from the University of Pennsylvania, and his BS and MS degrees from Wuhan Univ. and Rutgers University-New Brunswick, respectively. His research interests include machine learning, statistical modeling, and advanced data analytics, with application to data-enriched fields, especially bioinformatics. His research has been supported by the NIH, NSF, DoD, non-profit foundations, and industry companies. He is also the recipient of the Adobe Data Science Research Award. His methodological works have been published in premier machine learning and AI journals and conferences (Nature Machine Intelligence, Nature Communications, IEEE TNNLS, NIPS, KDD, AAAI, IJCAI, etc), and prestigious statistics/biostatistics journals (JASA, Biometrika, AOAS, Biostatistics, AJHG). Besides methodological research, Dr. Wei also enjoys collaborative research, which has resulted in many groundbreaking discoveries published in top scientific journals, including Nature, Science, Cell, Nature Medicine, Nature Genetics, Cancer Discovery, and Lancet Neurology, among others. In total, Dr. Wei has more than 200 publications in reputable journals and premier conferences with 17000 citations and an H-index of 59.