BMI Progress Reports 2021 // Dr. La Manno's Lab: A. Lederer - Intrinsically consistent RNA velocity estimation unveils cell cycle speed dynamics in neurodevelopment

Single-cell RNA sequencing (scRNA-seq) has transformed the study of complex biological systems by enabling gene expression measurements for thousands of individual cells. RNA velocity, using the unspliced and spliced RNA content captured during sequencing, introduced vector fields in a low dimensional space to infer a relative rate of change in gene expression. This opened the way towards temporal interpretations of biological processes from scRNA-seq. While the dimensionality of cellular differentiation variables is often high, the cell cycle, one of the main drivers of cell-to-cell heterogeneity, tends to unwind on lower-dimensional manifolds. We expand upon the concept of RNA velocity further by formulating the problem in terms of an autonomous dynamical system and inferring a consistent model for cell cycle dynamics on a periodic manifold. We validate our approach on simulated data and proliferating RPE1 FUCCI cells. Moreover, we apply our method to radial glial progenitors of the development mouse brain, revealing distinct proliferation modes in a regionalized and time-dependent manner. Together with multimodal omics methods, we hope to explore patterning of the anterior neural tube. These efforts will yield a powerful tool for the estimation of consistent RNA velocities and provide novel insight into nervous system development.