With recent technological developments it has become possible to characterize a single cell’s genome, epigenome, transcriptome and proteome. In particular, single-cell RNA-sequencing (scRNA-seq) has been widely applied to study heterogeneity in populations of neurons, in the immune system and in early development, revealing the existence of new populations of cells and differentiation trajectories. Fully exploiting such data requires the development of novel computational methods, with many of the tools developed for bulk RNA-sequencing not being appropriate for scRNA-seq. In this presentation I will describe some of the methods we have developed to address these challenges, and will illustrate their application in the context of immunology and early mammalian development.
I will begin by describing a method we have developed that distinguishes signal-from-noise in such datasets, and how we have used it to study differences in transcriptional heterogeneity in naive and activated CD4+ T cells from young and old mice. I will then describe how we used scRNA-seq to explore how differences in the affinity between a T Cell Receptor and its cognate peptide impact T cell activation. Finally, I will discuss new computational tools we have developed to handle the analysis of very large scRNA-seq datasets, which I will illustrate using data we have generated to study early mammalian development.