Identification of spatial patterns in spatially resolved multi-omics
No Thumbnail Available
Authors
Meeting name
Sponsors
Date
Journal Title
Format
Thesis
Subject
Abstract
Spatially resolved multi-omics enable the investigation of cellular and molecular interactions within their native tissue environments, offering unprecedented insights into tissue organization and function. Identifying spatially variable features (SVFs) is essential for understanding spatial gene expression patterns and regulatory mechanisms across multi-omics data. This dissertation introduces a series of methodologies designed for the detection and analysis of SVFs in spatially resolved transcriptomics and multi-omics data. First, we present BSP, a dimension-agnostic model that leverages spatial granularity to identify spatially variable genes across two- and three-dimensional datasets. Next, we extend this approach with scBSP, a highly efficient framework that scales to high-resolution spatial omics datasets using sparse matrix operations and approximate nearest neighbor search. Finally, we introduce a Statistical Tool for Analysis of Multi-omics Patterns (STAMPs) for the SVF detection on spatially resolved multi-omics that enables power analysis and implement cell type information. Overall, these advancements provide a robust analytical foundation for the identification of spatial patterns in spatially resolved multi-omics research, facilitating discoveries in tissue biology and disease mechanisms.
Table of Contents
DOI
PubMed ID
Degree
Ph. D.