Skip to main content
Fig. 6 | Genome Biology

Fig. 6

From: Systematic assessment of gene co-regulation within chromatin domains determines differentially active domains across human cancers

Fig. 6

Differentially active domains in normal vs. tumor prostate samples and cell lines. a Domain mean gene expression fold-change (X-axis) and domain mean gene expression correlation (Y-axis) for the comparison between prostate cancer and normal prostate tissue samples (Hi-C data from the lung cancer cell line RWPE1). Size of the dot is proportional to significance of the chromatin domain, blue (red) indicates negative (positive) average fold-change of significant domains. Highlighted are the top 4 chromatin domains determined by DADo. b, c Observed vs. expected Hi-C contact maps for genomic regions in chromosome 17 comprising the differentially active domain CD147 (b) and CD174 (c). The lower triangular contact maps correspond to the RWPE1 cell line, and the upper triangular contact maps correspond to the 22Rv1 cell line. The contact maps corresponding to the two domains are zoomed at the bottom to improve visibility. d, e Significant interactions estimated with the HiCDC algorithm in CD147 (d) and CD174 (e) in 22Rv1 (top) and RWPE1 (bottom) cell lines. HiC-DC p values < 0.1 (−log10(p) = 1) are color coded, white cells correspond to HiCDC p values > 0.1. ChIP-seq tracks for H3K27ac are shown below each map. f Gene expression values of the genes in CD147 (left) and CD174 (right) in RWPE1 (blue, 4 replicates) and 22Rv1 (red, 1 replicate). g Distribution of p values (-log10(p)) obtained from differential interactome analyses of Hi-C contacts within differentially active domains, separately shown for domains that were found more active in normal samples (blue) and in tumor samples (red). The sign of the −log10(p) values was set to positive (negative) for interactions that were more frequent in the tumor (normal) Hi-C dataset

Back to article page