Speaker: Pablo Gonzalez-Camara, UPenn

Title: Going Beyond Cell Type Identification with Single-Cell Transcriptomics

Abstract: Highly-parallelizable single-cell RNA-sequencing has emerged as a powerful technology for profiling the transcriptome of millions of individual cells within tissue samples. Analysis of scRNA-seq datasets by dimensional reduction, clustering, and differential expression analysis has led to the characterization of many cell populations, including rare populations. However, cells continuously respond to stimuli from their environment and their transcriptome is the result of many concurrent molecular pathways. Dissecting the transcriptional logic underlying cellular dynamics requires the development of new computational approaches that do not rely on clustering. In this talk, I will present two methods we have developed in this direction. In the first part, I will present a formalism for clustering-independent differential analyses of genomic data. Building upon spectral graph techniques, manifold learning, and topological data analysis, our formalism enables the study of transcriptional differences influenced by dynamic cellular processes and spatial context. In the second part of the talk, I will present a method for enriching highly-multiplexed immunostained slides with single-cell RNA-seq data. Our approach boosts the level of detail in histological analyses by enabling the detection of subtle cell populations, spatial patterns of transcription, and cell-to-cell interactions.