“Modeling the Dynamics of Genome-Scale Data Across Trees”
Recent efforts to elucidate molecular mechanisms of cellular differentiation and transitions to disease states have generated big data: thousands of genome-wide profiles of RNA expression together with DNA and histone modifications across hundreds of cell types. With the increasing availability of such information, the main roadblock to understanding how cells coordinate differentiation and maintenance is the difficulty of analyzing these large, complex datasets. The purpose of this grant is to develop statistical methods and software that enable genome-wide identification of epigenomic and gene expression changes that drive cellular transitions. Importantly, different cell types are related in a hierarchical manner as a result of their developmental trajectories. The central innovation of our approach is to leverage such relationships, which can be encoded in cell lineage trees, in our statistical models.