Fig. 1: CloneSelect.

a, Conceptual diagram of retrospective clone isolation. b, Different barcode-specific gRNA-dependent reporter activation circuits. CloneSelect C→T, CaTCH and ClonMapper. c, Barcode-dependent reporter activation of six barcoded cell lines by CloneSelect C→T. BC, barcode. Scale bar, 50 µm. d, Performance comparison of CloneSelect C→T, CaTCH and ClonMapper across the same set of barcode–gRNA pairs. Receiver operating characteristic curves were obtained by varying EGFP intensity thresholds for each target. The dashed line indicates the expected random classification. e, Percent positive cells with a uniform EGFP intensity gate applied to all of the tested systems (n = 3). Two-tailed Welch’s t-test was used for statistical analysis. f, Fold change between percent EGFP⁺ cells of OT (on-target) and NT (non-target) gRNA-barcode pairs for each barcoded cell line. Two-tailed Mann–Whitney U-test was used for statistical analysis. g, CloneSelect A→G. h, Barcode-dependent reporter activation of three barcoded cell lines by CloneSelect A→G. Scale bar, 50 µm. i, Performance comparison of CloneSelect A→G, CaTCH and ClonMapper across the same set of barcode–gRNA pairs. j, Percent positive cells with a uniform EGFP intensity gate applied to all of the tested systems (n = 3). Two-tailed Welch’s t-test was used for statistical analysis. k, Comparison of CloneSelect C→T and CloneSelect A→G. Activated cell frequencies of each barcoded cell sample by OT and NT gRNA queries were normalized by that of the cell sample with the same OT barcode–gRNA pair for CaTCH, ClonMapper or low-copy CRISPRa. The sample sizes for C→T barcodes and A→G barcodes were 18 and 9, respectively. Two-tailed Mann–Whitney U-test was used for statistical analysis; n.s., not significant; *P < 0.05; **P < 0.01; ***P < 0.001.