Skip to main content
Figure 6 | Epigenetics & Chromatin

Figure 6

From: DNA demethylation by 5-aza-2′-deoxycytidine is imprinted, targeted to euchromatin, and has limited transcriptional consequences

Figure 6

Limited transcriptional consequences of promoter demethylation by 5-aza-CdR. We show in panel (a) the RefSeq-defined HCG promoters and in panels (b) and (c) the RNAPII-Ser5 (P)-defined HCG and LCG promoters, respectively. On the left is a plot representing the k-means clustering of promoter methylation, with a very small proportion of promoters losing DNA methylation in both of the HCG categories (a, b) but a greater proportion of LCG promoters (c). When we take the genes associated with the k-means subgroups indicated by the colored percentages and represent their expression changes individually as gray lines and as subgroups depicted by colored lines, again reflecting the use of k-means clustering, we see very few genes to change expression levels, even though the proportion of promoters changing DNA methylation in panel (c) was relatively substantial. Percentages in the panels on the right represent proportions of all of the loci tested from the outset and are directly comparable with the percentages in the panels on the left. For example, while (23.0 + 3.7 + 5.4 =) 32.1% of the 1,338 promoters in panel (c) show loss of DNA methylation, when the expression of the associated genes is studied, only 0.45% of the 1,338 have induction of expression. LCG, low-CG dinucleotide observed/expected ratio; HCG, high-CG dinucleotide observed/expected ratio; RNAPII-Ser5 (P), serine 5-phosphorylated RNA polymerase II; FPKM, fragments per kilobase per million reads.

Back to article page