Skip to main content
. 2020 Nov;30(11):1605–1617. doi: 10.1101/gr.264283.120

Figure 6.

Figure 6.

Global H3K36me3 removal is associated with a one-phase exponential rate of decay. (A) Genome browser ChIP-seq signal track of a representative example of LANS-Set2 inactivation in a set2Δ (green) or set2Δrph1Δ (red) background over the time course experiment. Signal is normalized by internal spike-in S. pombe DNA and H3 ChIP-seq signal. (B) Distribution of mean, per-gene normalized LANS-Set2 inactivation ChIP-seq signal represented as interquartile box plots across the time course in a set2Δ (green) or set2Δrph1Δ (red) background. Each line represents the mean of all three replicates for one specific gene over time. Gray box plots represent LANS-Set2 inactivation in the set2Δ background (as seen in Fig. 3C), and black box plots represent the set2Δrph1Δ background. (C) Distribution of the relative LANS-Set2 inactivation ChIP-seq signal in a set2Δ (green) or set2Δrph1Δ (red) over time. To highlight relative H3K36me3 change for each gene, the maximum signal value over all time points was set to 1, and subsequent time points became a fraction of that maximum. Gray box plots represent LANS-Set2 inactivation in the set2Δ background (as seen in Fig. 3E), and black box plots represent the set2Δrph1Δ background. (D) Venn diagram of the high-confidence genes identified within LANS-Set2 inactivation in a set2Δ (green) or set2Δrph1Δ (red) background. High-confidence genes were determined to have a clear negative trend based on the 95% credible interval of the GLMM rate never including zero (on the linear predictor). (E) Histogram of the GLMM rates within the shared high-confidence gene set between LANS-Set2 inactivation in a set2Δ (green) or set2Δrph1Δ (red) background (n = 3531). (F) Jointly modeling the set2Δ and set2Δrph1Δ backgrounds of LANS-Set2 inactivation rates revealed a lag in set2Δrph1Δ compared to set2Δ across the time course. Points above zero indicate genes with a higher GLMM loss rates in set2Δ compared to set2Δrph1Δ at a given time point, whereas points below zero indicate genes with lower GLMM rates in set2Δrph1Δ compared to set2Δ. Each line represents one gene across the time course. Box plot borders represent the interquartile range of the difference in GLMM rates between set2Δrph1Δ and set2Δ for each time point.