Fig. 6

NSD3 deposits H3K36me2 specifically at active promoters and enhancers. a Genome-browser tracks displaying MS-normalized H3K36me2 signals where the distribution becomes progressively more punctate in multiple-KO conditions. b Genome-browser tracks showing NSD1/2-SETD2-TKO (TKO) H3K36me2 signal at active enhancer regions, which are identified by the presence of ATAC-seq and H3K27ac peaks. c Genome-browser tracks displaying H3K36me2 signal for TKO at active promoter regions, which are identified by the presence of ATAC-seq and H3K27ac peaks. d Overlap enrichment results of Ensembl annotations with bins found in TKO but not in QKO (i.e. identified as NSD3-catalyzed H3K36me2). The size of the dots corresponds to the number of bins overlapping the corresponding annotation. **** represents p-value < 1e − 4 whereas *** represents p-value < 1e − 3 based on Fisher’s exact test of bins overlapping a specific class of annotated regions versus a background of all bins in TKO. Only the top four most significant annotated regions are shown. e Heatmaps showing H3K36me2 (input- and depth-normalized as well as MS-scaled) signal centered on transcription start sites (TSS) in inactive (n = 18,290) and active promoters (n = 18,290) as well as H3K36me2 signal centered on inactive (n = 1758) and active enhancers (n = 1758), illustrating the loss of H3K36me2 signal following NSD3-KO. For a, b, c, and e, at least two replicates (n = 2) were merged for each ChIP-seq and ATAC-seq track. ChIP-seq signals were MS normalized and represent mean local frequency of the relevant modification. ATAC-seq signals were normalized by library depth