Skip to main content
Fig. 3 | Genome Biology

Fig. 3

From: Clipper: p-value-free FDR control on high-throughput data from two conditions

Fig. 3

Comparison of Clipper and popular bioinformatics methods in terms of FDR control and power. a peaking calling analysis on semi-synthetic ChIP-seq data. b Peptide identification on real proteomics data. c DEG analysis on semi-synthetic 10x Genomics scRNA-seq data. d DIR analysis on semi-synthetic Hi-C data. In all four panels, the target FDR threshold q ranges from 1 to 10%. In the “Actual FDR vs. Target FDR” plot of each panel, points above the dashed diagonal line indicate failed FDR control; when this happens, the power of the corresponding methods is not shown, including MASC2 and HOMER in a, edgeR in c, and multiHICcompare and FIND in d. In all four applications, Clipper controls the FDR while maintaining high power, demonstrating Clipper’s broad applicability in high-throughput data analyses

Back to article page