Figure 3

HmeDIP-SC-seq derived 5hmC data is comparable to published datasets using alternative profiling methodologies. (a) Scatter plots for 500,000 random 200 bp windows to plot biological replicate liver 5hmC levels for published hmeDIP tiled microarrays (i) or between hmeDIP-SC-seq (ii) datasets. R² values are shown in each plot. For i, plots refer to log² scores whilst ii refers to normalised average reads in a given 200 bp window. (b) Pearson’s correlation matrix with hierarchical clustering for replicate hmeDIP-SC-seq, input SC seq, hmeDIP-Hiseq, published affinity 5hmC Illumina Hiseq and 5hmC tiled microarray datasets. (c) Visual example of 5hmC patterns in the average input normalised hmeDIP-SC-seq (red), input normalised hmeDIP-Hiseq (green), published affinity Hiseq (blue) and published hmeDIP microarray datasets (gold) over a 2.5 Mb region on chromosome 7 (chr7:85,183,840–87,720,240). For a comparison, average input SC-seq data is shown in grey below. On the right, comparative 5hmC profiles are shown for hmeDIP-SC-seq and 5hmC affinity Hi-seq datasets over select loci (Cdkn1a and Klf3).