Fig. 5

Impact of imbalance ratios of cells among cell types on the performance of the number of cell type estimation. a Deviation of the estimated and the true number of cell types for each clustering method on datasets with different imbalance ratios (i.e. 2:1, 4:1, and 10:1) and different number of cell types (i.e. 10 and 20). Each combination was repeated 10 times for estimating variability. b Concordance of clustering output and predefined cell type labels as quantified by four concordance measures for datasets with different imbalance ratios of 2:1, 4:1, and 10:1, and error bars represent the standard deviation