Fig. 5: Application of analytical framework to breast cancer cohort.
From: Chromatin accessibility associates with protein-RNA correlation in human cancer
a Cohort characteristics, includes 19 breast cancer (BRCA) samples (11 luminal B, 4 luminal A, and 4 basal). b Differential transcription factor footprints between luminal A/B and basal subtypes. c Visualization of gene accessibility and transcript and protein abundances for EGFR in BRCA subtypes. Biologically independent tissue chunks from each disease group (i.e., lumA, lumB, and basal) are averaged to produce the signal plots. The gray highlighted region on the signal track indicates footprinted differentially active enhancers d Density of sample-wise correlation between gene accessibility and expression for not differentially active genes (top) compared to differential active genes (bottom). e Heatmaps of gene accessibility (top) and protein abundance (bottom) for genes that are differentially expressed and contain gene body enhancers. Annotation bars delineate samples’ subtypes (top bar) and gene membership in known breast cancer gene sets (left bar). f Protein–protein interaction networks for luminal regulons (top) and basal regulon (bottom). Green dots mark known PAM50 genes.
