Extended Data Fig. 7: Additional data of regional analyses and clustering of clones at E11.5.
a, Visualization of dispersion behaviors of clones along the dorsal-ventral axis in the E11.5 dataset, where circles, sectors, and boxes are colored by regional identities as described in Fig. 4b. b, Distribution of dispersion scores in E11.5 clones. c, Relationships between dispersion score and maximum percentage (calculated as [number of cells with the most dominant regional identity in each clone] / [total number of cells in each clone (referred to as clone size)]); between clone size and cell type complexity (calculated as the Shannon-Wiener Index based on the clonal cell type compositions); and between clone size and dispersion score of clones in E11.5 (upper) and E15.5 (lower) datasets. d, e, Clustering of pairwise lineage correlations as determined by CREST barcodes shared between midbrain domains in the E11.5 dataset (d) and the E15.5 dataset (e). Cell types with a same regional identity in each dataset (as described in Fig. 2f–j) were merged into a ___domain cluster. For example, NbFP and Rgl1 in the E11.5 dataset were merged as the E11.5 floor plate (FP) cluster. Cell types with undefined regional identities were discarded. f, Quantification of the number of clonal cell types in clones from the E11.5 (left) and E15.5 (right) datasets. Multipotent progenitors (producing at least two distinct cell types) were predominant, giving rise to 85.8% and 88.8% of individual clones at E11.5 and E15.5, respectively. g, Clones in the E11.5 dataset are clustered based on the cell type composition matrix. h, Visualization of E11.5 clone clusters mapped on the UMAP space. The UMAP reference is colored based on spatial identities.
