Modeling cellular state as well as dynamics e.g. during differentiation or in response to perturbations is a central goal of computational biology. Single-cell technologies now give us easy and large-scale access to state observations on the transcriptomic and more recently also epigenomic level. In particular, they allow resolving potential heterogeneities due to asynchronicity of differentiating or responding cells, and profiles across multiple conditions such as time points, space and replicates are being generated.
In this talk I will shortly review scVelo, our recent model for dynamic RNA velocity, allowing estimation of gene-specific transcription and splicing rates, and illustrate its use to estimate a shared latent time in pancreatic endocrinogenesis. I will then show CellRank, a probabilistic model based on Markov chains which makes use of both transcriptomic similarities as well as RNA velocity to infer developmental start- and endpoints and assign lineages in a probabilistic manner. It allows users to gain insights into the timing of endocrine lineage commitment and recapitulates gene expression trends towards developmental endpoints.
While the above approaches focus on individual gene expression models, recently latent space modeling and manifold learning have become a popular tool to learn overall variation in single cell gene expression. I will wrap by briefly discussing how these tools can be used to integrate single cell RNA-seq data sets across multiple labs in a privacy aware manner.