Extended Data Fig. 7: Reproducibility of the CGE experiments.

a, b, Internal replicate analysis performed by splitting individual cell lineages harboring unique sequence tags to different bins (see Methods). In a, cell lineages harboring mutations at the E-box locations of the MDN1 promoter are binned on the basis of the mutated positions in the 10-bp sequence tag so that the tags having the first flanking mutation at an odd and even positions are separated to two bins, respectively. In b, cell lineages harboring CDK1 T14A + Y15F mutations are binned to four groups on the basis of the randomized nucleotides in the sequence tag so that the sequence tags having A, T, G, or C as their first mutated nucleotide are distributed to different groups, respectively. c, The fitness effects of E-box mutations from two independent experiments, demonstrating the reproducibility of the results from the CGE method. In panels a-c, each dot represents an individual cell lineage harboring a unique barcode. Log₂ values for read count ratios (day 8/day 2) are shown for each sequence tag pair after calculating the ratio of read counts for mutated vs. original sequence at each time point. Red lines represent the median values; p-values from two-sided Wilcoxon signed rank test are shown for each experiment (no multiple comparison adjustments; see Supplementary Table 3 for details of statistical parameters).