Fig. 5

Genic H3K36me2 distribution patterns are altered following SETD2-KO, NSD1/2-DKO, and H3K36M-OE. a Genome-browser tracks of MS-normalized H3K36me2 signal (merged replicates, n = 3 per condition) showing changes in genic distribution following specific K36MT knockouts. b Zoomed in aggregate view of H3K36me2 distribution in gene bodies across conditions, highlighting changes in exonic versus intronic signal and dependence on gene expression. Transcripts with at least 50,000 bp and 6 exons were used. An aggregate of MS-normalized H3K36me2 signals (merged replicates per condition, n = 3) from the first three exons and the last three exons are shown. Expression quantiles were calculated based on normalized expression counts from an average of parental replicates. Expression quantile 4 comprises transcripts with the highest expression whereas expression quantile 0 comprises transcripts with zero counts. c Correlation plots of depth-normalized H3K36me2 signal and gene expression quantiles, indicating a shift to significant positive correlations between H3K36me2 and gene expression following SETD2-KO, NSD1/2-DKO, and H3K36M-OE. Pearson’s correlation coefficient (Pearson r) is given as well as the p-value of the linear correlation, which indicates whether the linear correlation between H3K36me2 signal within gene bodies and gene expression quantiles is significant for a given sample. ChIP-seq signals were normalized by dividing by the total alignments (CPM) to clearly visualize the differences in correlation across cell lines with gene expression and not intended to indicate differences in total H3K36me2 abundances. d Boxplots of mean H3K36me2 signal at exons versus introns, indicating a trend towards greater signal at exons compared to introns following SETD2-KO, NSD12-DKO, and H3K36M-OE. Boxes span the lower (first quartile) and upper quartiles (third quartile), median is indicated with a center line and whiskers extend to a maximum of 1.5 times the interquartile range