报告人: 吴梦云（上海财经大学）
报告时间: 5月20日 19:00-20:00
报告地点:#腾讯会议：929-366-038
Abstract: Identifying genes that display spatial patterns is critical to investigating expression interactions within spatial context and further dissecting biological understanding of the complex mechanistic functionality. Despite the increasing number of statistical methods designed to identify spatially variable genes, they are mostly based on marginal analysis to model each gene separately and share the limitation that the dependence (network) structures among genes are not well accommodated, where a biological process usually involves changes in multiple genes that interact in a complex network. Moreover, the latent cellular composition within spots may introduce confounding variations, resulting in poor identification accuracy. In this study, we develop a novel Bayesian regularization approach for spatial transcriptomic data, with the confounding variations induced by varying cellular distribution being effectively corrected. Significantly advancing from the existing studies, a specified thresholded graph Laplacian regularization is proposed to simultaneously identify spatially variable genes and accommodate network structure among genes. The proposed method is based on a zero-inflated negative binomial distribution, effectively accommodating the count nature, zero inflation, and overdispersion of the spatial transcriptomic data. Extensive simulations and the application to real data demonstrate the competitive performance of the proposed method over existing alternatives.
简介：
吴梦云，上海财经大学统计与管理学院，教授。中山大学理学博士，耶鲁大学生物统计系博士后。主要研究方向为高维数据统计分析方法及其在生物医疗领域的应用。在国际统计学及生物统计学重要期刊Annals of Applied Statistics、Statistica Sinica、Biometrics、Biostatistics、Genome Biology、Bioinformatics等发表论文40余篇。入选上海市晨光计划、上海市浦江人才、上海市启明星计划；主持国家自然科学青年基金项目和面上项目各一项；主持全国统计科学研究重大项目，并获得优秀结项。
邀请人：徐礼柏