Decoding cell division history and lineage-resolved phenotypic patterns from single-cell barcode and transcriptomic data

The cell division tree encodes how proliferation and differentiation generate cellular diversity during multicellular development. Advances in single-cell RNA sequencing and CRISPR-based lineage tracing have enabled retrospective reconstruction of these histories at single-cell resolution. Here, we present FateScape, a statistical framework that integrates paired lineage barcodes and transcriptomic profiles to infer cell division topology and characterize depth-resolved phenotypic patterns. FateScape combines barcode consistency with transcriptome-derived state continuity through overlapping state-lineage decomposition and barcode-guided subtree integration. Here, to analyze phenotypic patterns on the inferred tree, we introduce entropy paths, which measure how cells of each state are distributed across depth-defined subtrees, and Moran's I –derived statistics for same-state autocorrelation and cross-state association. In simulations, FateScape shows robust reconstruction performance across varying mutation rates, dropout levels, sample sizes, and barcode target-site numbers. In the Caenorhabditis elegans data analysis, FateScape accurately recovers lineage topology and identifies depth-resolved state patterns, including concentrated intestinal cells and broadly distributed neuronal and glial states. In mouse embryos, FateScape quantifies distinct germ-layer patterns, including early endodermal concentration, broad ectodermal distribution across shallow-to-intermediate depths, and intermediate mesodermal distribution followed by deeper concentration. Together, FateScape provides a framework for reconstructing cell division histories and quantifying state dispersion, concentration, and tree-based association across lineage depth.