Fig. 4

DNA methylation on gene bodies correlates with gene expression. A Heatmap of DNA methylation levels detected in Supra E brain samples across full-length transcripts and 2000 bp upstream and downstream of start site. All transcripts have been divided in 4 clusters by k-means based on the DNA methylation average across each transcript. B Line plot represents average of DNA methylation of Supra E and Sub E brain and hatchling for each cluster defined in the heatmap on Supra E sample. C Violin plot displays the distribution of transcript expression values (as log2(TPM+1)) for each methylation pattern in Supra E brain samples. Box-and-whisker plots inside the violin have a center line at the median, lower and upper hinges correspond to first and third quartiles, and whiskers extend from hinges to largest or smallest values no further than 1.5 × IQR (inter-quartile range), while data beyond the end of the whiskers are outlying points that are plotted individually. p-values were calculated by unpaired non-parametric Kruskal-Wallis test adjusted with Dunn’s multiple comparisons test. **** indicates adjusted p-value < 0.0001, and n.s. indicates not significant adjusted p-values > 0.9999. D Overall DNA methylation of transcripts (TSS to TTS of each transcript) in Supra E brain is regressed against transcripts expression (log2(TPM+1)). Transcripts were grouped by percentile of expression values and each dot represents the average value of DNA methylation for each percentile. E The expression profile (as log2(TPM+1)) of transcripts categorized in each methylation pattern have been represented across the 13 different tissues. Column clustering is supervised according to the sample order established in Fig. 1B. Row clustering is calculated on the hierarchical clustering of dendrogram (Euclidean distance). F Gene Ontology (GO Biological Process) annotation of transcripts in each methylation pattern. Dot size represents gene ratio between observed and expected transcripts in each GO category (with padj < 0.05), dot colors represent adjusted p-value