Extended Data Fig. 7: cDNA barcode amplification and recovery.

(a) Schematic diagram demonstrating principle of three amplification strategies. (b) Step-by-step statistics of cDNA barcode recovery. Data were separately counted in TT, TN, or NN amplicons. (c) Number of CCS reads assigned to an original barcode sequence. Each dot represents a UMI-consensus 1-kb barcode sequence. Data were separately counted in TT, TN, or NN amplicons. All quality barcode sequences were counted. (d) Number of UMI-consensus 1-kb barcode sequence assigned to a single cell. Each dot represents a single cell. Data were separately counted in TT, TN, or NN amplicons. All quality cells were counted. (e) Number of UMI-consensus 1-kb barcode sequence recovered per cell. All quality cells were counted. UMI generated from TT, TN, or NN amplicons were integrated by cell. Box plots in (c,d,e) show the median (centre line), the 25th and 75th percentiles (box bounds), and whiskers extending to the most extreme values within 1.5× the interquartile range. Notches indicate the approximate 95% confidence interval of the median. Outliers are not shown. (f) Correlation between barcode expression level and barcode recovery rate. Barcode expression is counted by the 1-kb barcode sequence detected in scRNA-seq. Barcode recovery rate indicates the number of full-length cDNA barcode recovered within a cell. (g) UMAP showing the number and percentage of cells with quality lineage barcode recovered in each library. (h) Number of cells with quality lineage barcode recovered in each plant and library. (i) Mutation spectrum of cDNA barcode. (j) Base editing window of cDNA barcode.