Abstract: The emergence of single-cell lineage tracing technologies has enabled the reconstruction of phylogenetic trees for thousands of cells, facilitating the application of phylodynamicsinference (PI) at the cellular level. However, the complexity of cell differentiation presents significant challenges for existing PI frameworks. To address these challenges, we present scPhyloX, a novel computational approach that utilizes single-cell phylogenetic trees to infer dynamics of tissue development and tumor evolution. Simulations demonstrate that scPhyloXachieves high accuracy, while analyses of real datasets provide new insights into somatic dynamics, such as stem cell population overshoot during fly organ development and clonal expansion in human aging and early colorectal tumorigenesis. Concurrently, single-cell RNA sequencing (scRNA-seq) is a powerful tool for investigating cellular differentiation; however, tracking cell fate transitions in disease contexts remains difficult. We introduce PhyloVelo, a framework that estimates transcriptomic dynamics using monotonically expressed genes (MEGs). By integrating scRNA-seq data with lineage information, PhyloVeloreconstructs a transcriptomic velocity field. Validation studies demonstrate that PhyloVeloaccurately recovers differentiation trajectories, surpassing traditional RNA velocity methods, and identifies MEGs with conserved functions in translation and ribosome biogenesis across various tissues and organisms.
