Systems Biology Group Meeting

Single-cell RNA-seq and single-cell ATAC-seq reveal gene expression and chromatin accessibility of individual cells. In steady-state populations, this granularity allows us to explore distinct subpopulations of cells and the regulatory mechanisms that differentiate them. In dynamic processes, single-cell resolution provides a snapshot of cells as they undergo biological processes. As these genomic technologies improve, insight into regulatory mechanisms that drive dynamic processes and cellular diversity requires new computational approaches.

 In this talk, I will present my work identifying regulatory chromatin implicated in the dynamic oligodendrocyte progenitor cell (OPC) to oligodendrocyte transition in human spinal tissue. In this same dataset, I will discuss future directions for network inference with Variational Autoencoders (VAEs) in steady-state Astrocyte populations across the rostro-caudal axis. Finally, I show we can learn stochastic ODEs that govern dynamic processes on published data. I will highlight how inferred “drift” and “diffusion” fields in expression space can open new avenues in how we approach gene regulatory networks outside steady-state assumptions.