Measuring and modeling gene expression trajectories: new computational-experimental approaches

Boston University (Charles River Campus), Boston MA

Investigators

Abstract

RNA production, processing, and degradation are tightly regulated to articulate changes underlying a wide array of biological processes that are central to development, function in adult tissue, and disease progression. RNA kinetic rates jointly determine the “velocity” of each gene, which describes the rate of change in transcript abundance. Studying RNA velocity at the single cell level allows one to access variability – either natural heterogeneity or through pooled single cell perturbation screens – that can be used to uncover regulatory associations by observing how velocity and kinetic rates change when the cell profile is altered. Analysis of velocity profiles across a population of cells can in turn paint a far richer, potentially transformative picture of cell states: in observing the dynamic profile of many single cells in a population, one can, in principle, piece together observed short-term changes in each cell and map long-term expression trajectories that were never directly observed, charting the paths that single cells take in dynamic processes and providing a foundation for understanding the phenotypic space that cells can occupy. The overarching goal of work in our lab is to make foundational advances enabling the study of single cell gene expression trajectories. To do so, we utilize an organically interdisciplinary approach that is grounded in experimental Systems Biology and Genomics, but that draws heavily from Machine Learning and Data Science both for computational analysis and to directly inform experimental design. Our work is unified and distinguished by our deep commitment to integrating computation and the principles of data science with biological experimentation in a wholistic way. The proposed research pursues three distinct but conceptually interrelated research directions: (1) To develop new experimental strategies that track RNA produced in single cells at multiple timepoints, allowing for more accurate estimation of RNA velocity; the discovery of patterns of joint variation in RNA production, degradation, and expression levels; and enabling studies that aim to learn the regulatory circuitry of RNA degradation at scale. (2) To learn causal RNA velocity vector fields by integrating single cell RNA velocity measurements with high-throughput genetic perturbation screens, allowing us to find a vector field map that accurately captures the trajectory landscape and leads to new biological insight by identifying genetic programs that guide trajectories. (3) To establish a systematic, quantitative understanding of how cell physiology impacts the “total RNA velocity” of a cell by observing the total RNA velocity (including the 97% of cellular RNA that is ignored in mRNA studies) in single cells under different physiological conditions, interventions to modulate a cell’s total velocity, and the functional consequences of a total velocity limit. Collectively, these parallel and complementary research directions focusing on single cell measurements, population-level inference, and the physiological limits of coordinated changes in expression will establish concrete methods and conceptual frameworks that will enable scientists tackling a wide range of fundamental problems relevant to human disease.