Skip to main content
Fig. 3 | BMC Bioinformatics

Fig. 3

From: A comparative study of topology-based pathway enrichment analysis methods

Fig. 3

Average ranking of empirical powers on the 11 KEGG (primarily signaling) pathways using sample labels from the original study (a) and shuffled sample labels (b) based on the betweenness dysregulation design for the TCGA breast cancer study [33]. The x-axis shows the pathway size, and the y-axis indicates the average ranking of empirical power over different mean changes. Lower ranking indicates better performance. Overlapping points were re-positioned by adding ± 1 to the x-axis and ± 0.1 to the y-axis. PathNet, CAMERA and PE.noCut perform the best when using the original sample labels (a), whereas DEGraph, PE.noCut and topologyGSA yield the best performance with permuted sample labels (b)

Back to article page