Fig. 4: Large-scale new RNA analysis of epigenetic compounds effects in K562 cells. | Nature Communications

Fig. 4: Large-scale new RNA analysis of epigenetic compounds effects in K562 cells.

From: Deciphering direct transcriptional effects of epigenetic compounds through large-scale new RNA profiling

Fig. 4: Large-scale new RNA analysis of epigenetic compounds effects in K562 cells.The alternative text for this image may have been generated using AI.

a Boxplots showing the number of significantly up- and down-regulated genes per compound, grouped by mode of action. Boxplots show median, boundaries (first and third quartile), and whiskers denote 1.5 times the interquartile range of the box. Sample-sizes per group (HDAC inhibitors: 28, SIRT activators: 5, HAT inhibitors: 5, HAT activators: 2, HDME inhibitors: 6, MET inhibitors: 17, Bromodomain inhibitors: 14, other: 3). b Principal components and the percentage of variance explained by each component for the analysis of all 83 epigenetic compounds. c Heatmap showing the scores of each compound onto principal components 1 to 10, where compounds (rows) are colored according to their overall mode of action. d Heatmap of average new RNA differences for significantly up- or down-regulated genes, comparing each bromodomain inhibitor against DMSO-treated control cells (three biological replicates). Both compounds (columns) and genes (rows) where subject to hierarchical clustering. e, f Volcano plots showing enrichment of transcription factor binding (ChIP-seq) in the promoters of genes in clusters C1 (e) and C4 (f), compared to the union of genes in the other clusters. g Heatmap with average new RNA differences for significantly up- or down-regulated genes (as in d) comparing each DNA methylase inhibitor against DMSO-treated control cells. (h,i) Volcano plots showing factor binding enrichment in the promoters of genes in clusters C2 (h) and C3 (i), respectively. e, f, h, i Enrichments were evaluated using Fisher’s exact test (two-sided), with the resulting P values adjusted using the Benjamini-Hochberg procedure. j Average linkage hierarchical clustering of breast cancer MCF7 compound responses on fitted dose-response slopes for 3479 genes. Colors indicate correlation strength (r = 0.33 cutoff), highlighting the PI3K/AKT/mTOR inhibitor class. Drug targets are shown in parentheses. k Significant gene ontology enrichment of compound-responsive genes identified from new RNA profiling of MCF7 cells. Downregulated genes were compared to expression-matched genes or upregulated genes, with P-values corrected for multiple testing within each of the resulting in 128 comparisons for 40 compounds that passed quality-control. Source data are provided as a Source Data file.

Back to article page