Skip to main content
. 2021 Jul 22;49(18):e104. doi: 10.1093/nar/gkab601

Figure 2.

Figure 2.

Comparison of the performances of the four methods with common gene sets from the glioblastoma scRNA-seq dataset. r0.1 represents the synthetic dataset in which 90Inline graphic of read counts present zero expression compared to the original data. The average AUROC (A), AUPR (B) and AP (C) were calculated by repeating each simulation 10 times.