Inducible genes have higher levels of the elongation mark H3K36me3. (a) Data from ChIP-Seq experiments on human CD4+ lymphocytes were used to determine the levels of H3K36me3 within the gene (+6 to +8 kb) for primary (red), secondary (blue) and unchanged (white) genes in each basal expression bin (Log2 robust multichip average values from expression profiling). The bar marks the median score, the edges of the boxes the second and third interquartile ranges and the whiskers the first and fourth interquartile ranges. (b) The genes with tag counts equal to or greater than the median level (14) for the unchanged genes in the basal expression Log2 6 to 7 bin were considered to have H3K36me3, and the percentage of genes that were H3K36me3 positive for each subgroup is shown. (c) From the same data source the number of sequencing tags for Pol II (-0.25 kb to +0.25 kb) and the putative elongation marks H2BK5me1 (0 to +2 kb) and H4K20me1 (0 to +7.5 kb) as well as H3K36me3 were counted for genes with basal expression values between Log2 3 and 6. The logs of the sequence counts were median centered and normalized and heatmaps for the primary response genes were generated by uncentered correlation, complete linkage clustering. The major clusters are marked and the genes are colored according to their basal expression level (green, log2 3 to 4; black, log2 4 to; red, log2 5 to 6). In the cluster diagram green indicates low tag counts and red indicates high tag counts. (d) ChIP assays were performed with antibodies against trimethylated H3K36 (H3K36me3) using unstimulated EL-4 T cells and detected by real-time PCR analysis. The data are presented as the ratio of immunoprecipitated DNA to the total input DNA and shows H3K36me3 occupancy at the promoter (green bars) and 2 kb downstream of the promoter (black bars). The mean and standard error of three independent experiments are shown.