Supplementary Fig. 3: Association of chromatin interactions and the occupancy of CTCF, SMC1, YY1 and H3k4me1. | Nature Cell Biology

Supplementary Fig. 3: Association of chromatin interactions and the occupancy of CTCF, SMC1, YY1 and H3k4me1.

From: Epigenetic modulation of a hardwired 3D chromatin landscape in two naive states of pluripotency

Supplementary Fig. 3

a) Representative examples of chromatin interactions for a gene upregulated in 2i-ESCs (Ablim2) and a gene upregulated in serum-ESCs (c-Myc). Promoter-mediated cHiC interactions are shown in blue. 4C-seq profiles display the running median of 7 DpnII fragments. Dots represent the mean and error bars show the SEM based on n=4 biological replicates in serum-ESCs, 2i-ESCs and EpiLSCs. Fold-changes and significance were determined in bins of 5 kb, 1 Mb up- and downstream of the viewpoint. Differential interactions consistent between replicates are marked with asterisks (p < 0.1, * p < 0.05, ** p < 0.001, *** p < 1e-5; Two sided Wald test (DEseq2)). The viewpoint position is marked with an eye symbol. b) Normalized promoter cHiC interaction intensities of 10,211 genes in serum-ESC and EpiSCs (see text). The majority of the genes (85%) have no differential interactions. The 250 genes most up- or downregulated in the serum-to-2i transition also display few differential interactions between serum-ESCs and EpiSCs. Moreover, most differential interactions are found at promoters associated with H3K27me3 in serum-ESCs. c) Pearson correlation between biological duplicates employed in ChIP-seq analysis. Numbers in the graph indicate Pearson correlation coefficients. n=41630 CTCF_peaks, 35873 SMC1_peaks,19924 YY1_peaks. d) Change in H3K27ac level across interaction-anchors does not correlate with change in cHiC interactions. n=8,016 differential E2 peaks (involving 56,451 significant pairwise interactions), R-value represents Pearson’s correlation coefficient. e) Violin plot showing the occupancy of CTCF, SMC1, YY1 and H3k4me1 across differential enhancers that strongly change H3K27ac between serum- and 2i-ESCs (n=4,303 2i-specific E2 enhancers). Violin=kernel density estimation; center=median; bar=interquartile range; thin line=95% confidence interval. f) Interaction frequency between strongly differential (fold-change >= 3) H3K4me1 sites in 2i- vs serum ESCs (left) and serum-ESCs vs EpiSCs (right). Change in H3K4me1 is significantly associated with change of interaction intensity (One-sided Wilcoxon signed-rank test). Few promoters- and enhancers (n=236 enhancers) show differential H3K4me1 between serum- and 2i-ESCs, and this number is significantly higher (n=399 enhancers) when ESCs are compared with EpiSCs. P-values from left to right: P=0.03; P=0.008; P=7.2e-6; P=2.6e-9, Wilcoxon signed-rank test (one-sided); Box=5-75th percentile; bar=median; whiskers=5-95th percentile. n numbers represent enhancers and are stated in the graph for each group.

Back to article page