Figure 3
From: GSVA: gene set variation analysis for microarray and RNA-Seq data
Comparison of differential pathway activity identification of GSVA, PLAGE, single sample GSEA (ssGSEA) and combined z-score (zscore). Each panel shows the area under the ROC curve (AUC) on the y-axis for differentially expressed genes predicted by each method at 5% FDR over 100 simulations (see Methods). On top of each boxplot the p-value of the t-test for no difference in means between GSVA and the corresponding method is reported. The two panels on top correspond to simulations where 50% of the genes in DE gene sets were DE while the two at the bottom contained 80% of DE genes on those DE gene sets. The two panels on the left correspond to a weak signal-to-noise ratio in the DE magnitude while the two on the right correspond to a strong one. Diamonds indicate mean values in boxplots.