Fig. 3

Evaluate the performance of negative-defining methods using dual-platform single-cell sequencing data. Two negative-defining methods, UMC and MSN, were tested in a single-cell dual-platform whole-exome sequencing dataset using varying thresholds, including UMC (p < 0.01, p < 0.05) and MSN (minimum coverage of 10X, 20X, 50X, 100X, 300X and 100X), as indicated under each dot. The overall performance of each method with a specific threshold was evaluated by (1) Y-axis: the total number of informative data points after excluding unknown statuses, with “informative data points” defined as the single-cell/mutation pairs where both platforms (NXT and AGL) yielded an informative mutation status, i.e., either positive or negative, but not unknown; (2) X-axis: the concordance of mutation statuses between the two platforms (NXT and AGL), defined as the percentage of informative data points where the two platforms yielded the same mutation status, either both positive or both negative, for the same single cell