Fig. 4

Catalogue of 5,104 bivalent gene promoters in mouse embryonic stem cells. A Schematic outlining classification strategy for bivalent promoters. Full list of bivalent promoter classifications is available in Additional file 3: Table S3. B Overlap of all bivalent promoters (top) and high-confidence bivalent promoters (bottom) between those identified in this study compared to previously published reChIP data (reanalyzed from [14]). C Enrichment heatmaps showing CPM/bp normalised read densities for high confidence (top row), K4-biased (second row), K27-biased (third row) and low confidence (bottom row) bivalent promoters after scaling for all datasets analysed. Peaks were extended to 5kb upstream and downstream of the peak centre. Values surpassing the 99th percentile have been masked for visualisation. 10⁷ samples refers to independent total H3K4me3 and total H3K27me3 datasets from GSE135841 [7] D, E Scatterplot showing log₂enrichment (CPM/bp) of D in-line total H3K4me3 (x-axis) and in-line total H3K27me3 (y-axis) or E bivalent K4-K27 (x-axis) and K27-K4 (y-axis) reChIP datasets for all promoters highlighting those that overlap different classes of bivalent peaks defined using independent 10 million cell total H3K4me3 and total H3K27me3. F Box plot showing log₂ RPKM gene expression levels in mouse embryonic stem cells for four different classes of bivalent genes and those redefined from previous data (14). Expression of the bottom 20% and top 20% is shown as a comparison. Gene expression data reanalyzed from GSE135841. G Gene Ontology analysis showing overlap of representative enriched terms in the four classes of bivalent genes (top) and gene ratios and adjusted P-value of selected terms (bottom). The full list of enriched terms is available in Additional file 3: Table S4. H log₂ fold change in gene expression levels for different classes of bivalent genes across 9 days of embryoid body differentiation. Each gene has been normalised separately across the time series, genes are grouped using correlation based clustering. Gene expression data reanalyzed from (GSE135841)