Fig. 2From: A multi-omics approach for identifying important pathways and genes in human cancerVisualization of model predictive performance. Illustration of results from the analysis of TCGA pan cancer data using the MSigDB curated canonical pathways (C2.CP) collection. Rows correspond to the TCGA cohorts and columns to MSigDB gene sets. Each cell represents a pathway and cancer type-specific model colored according to the \(R^{2}_{pred}\) value. Both the columns and rows are ordered according to the output from hierarchical clusteringBack to article page