Fig. 6From: Cancer prognosis prediction using somatic point mutation and copy number variation data: a comparison of gene-level and pathway-based modelsDistribution of concordance index values for all models evaluated on the LGG cohort. “PLv” represents “Pathway-level” and “GLv” represents “Gene-level”. HALLMARK, PID, BIOCARTA, REACTOME and BP are the used five pathway collections as introduced in "Methods" section. Corresponding to each model introduced in "Methods" section: “log OR” and “Binary” represent the two enrichment methods for SPM data; “all genes”, “Path intersected”, “Cosmic intersected”, “Cox filter” represents the filters on the genesBack to article page