蜜桃传媒app

From Spatial Coordinates to Gene Discovery: Methodological Explorations in Spatial Transcriptomics

Qihuang Zhang, McGill University
Tuesday October 1, 12-1pm
Zoom Link:听
In Person: 550 Sherbrooke, Room 189

Abstract:听Spatial transcriptomics data have revolutionized the profiling of gene expression by incorporating spatial context, which offers a unique opportunity to study both the spatial organization of cells and their gene expression within tissues simultaneously. In this talk, I will provide an overview of statistical and machine learning methods that enhance our ability to extract meaningful insights from spatial transcriptomics data. Specifically, I will present approaches that leverage the different aspects of spatial locations and gene expression to address two key problems: cell mapping and the identification of important genes. First, I will discuss methods that focus on spatial location information to facilitate cell mapping. In particular, I will introduce CeLEry, a machine-learning method designed to predict spatial coordinates for single-cell RNA-seq data using spatial transcriptomics as a reference. This method improves our understanding of cellular arrangement within tissues and its relation to Alzheimer鈥檚 disease progression. The second approach focuses on gene identification, where we employ Bayesian hierarchical models to identify genes specific to disease status or associated with high cell densities. By identifying these key genes, we gain valuable insights into the biological mechanisms driving disease progression.