Skip to main content
. 2020 Jun 4;48(13):7154–7168. doi: 10.1093/nar/gkaa474

Figure 3.

Figure 3.

H2Bub1 levels are proportional to pSpt5 and correlated with antisense regulation by Cdk9 and RNAPII promoter-proximal pausing. (A) Top: Scaled metagene representations of pSpt5/Spt5 or H2Bub1/H2B ratios from transcription start site (TSS) to cleavage and polyadenylation site (CPS) for all genes. Genes were clustered based on total pSpt5/Spt5 occupancy. Bottom: Heat map representation of pSpt5/Spt5 and H2Bub1/H2B ratios for genes in each cluster. (B) Top: Metagene plots show the distribution of RNAPII, Spt5, pSpt5, H2B, and H2Bub1 around the TSS and CPS (±500 bp) for n = 3329. Bottom: Plots show the distribution of log2-ratio of pSpt5/Spt5 and H2Bub1/H2B around the TSS and CPS (±500 bp) for the same set of genes (n = 3329). (C) Metagene representations of H2Bub1/H2B or pSpt5/Spt5 ratios for the indicated groups of genes, plotted from the indicated distances (in kilobases) relative to the TSS. (D) Gene ontology (GO) terms enriched among genes in the ‘cdk9as antisense UP group’ (Fisher's exact test with Bonferroni correction). (E) Venn diagrams indicating overlap between the indicated gene groups and genes with promoter-proximal RNAPII pausing (hypergeometric test).