Fig. 4

Comparison of scMayoMap and other cell annotation tools on 48 real datasets. A Comparison of cell type annotation accuracy in PBMC datasets. Thirteen PBMC scRNA-seq datasets generated by seven sequencing technologies were tested, including two low-throughput plate-based methods (Smart-seq2 and CEL-Seq2) and five high-throughput methods (10 × Chromium, Drop-seq, Seq-Well, and inDrops). B Comparison of cell type annotation accuracy on 35 datasets with different tissues from Tabula Muris and literature. The accuracy of each method is presented as the percentage of cells that were correctly annotated by each method. C and D F1-score calculated for data in A and B. Jitters on the plot represent datasets. The whisker extends from the hinge to the value that is within 1.5 * interquartile range (IQR) of the hinge, where IQR is the inter-quartile range or distance between the first and third quartiles