Figure 2

Alu and LINE-1 repeats exhibit methylation changes upon CGGBP1-depletion. A: Mean methylation increase on Alu repeats measured by CG frequency per PCR product sequence. Y-axis shows nucleotide frequency calculated per sequence. X-axis shows the samples and treatments. B: Frequency distribution of Alu methylation across different ranges shows decrease (<8%) and increase (>12%) at extremes in CGGBP1 shmiR sample as compared to Control shmiR sample. Y-axis shows relative frequencies of CG (a measure of methylation; normalized for different number of sequences per sample). C: Plotting of the >12% and <8% methylation subset from B shows the tailing of differentially methylated sequences at both extremes clearly. D: Frequency plot of data plotted in C and best curve fit shows sum-of-two-Gaussian fit for CGGBP1-depleted sample and a single Gaussian distribution for control sample suggesting that indeed there are two groups of methylation levels for Alus in CGGBP1 shmiR but only one group of methylation level in Control shmiR sample. E and F: Increase in CpG content negatively correlated with TpG frequency in both samples establishing the fact that the changes in cytosine content was indeed due to bisulfite conversion of unmethylated cytosines. G: Increase in methylation on LINE-1 elements was significant with no bidirectional heterogeneity as seen for the Alus. H: Frequency plotting showed that <7.5% (marked with dotted line) methylation was prevalent in control sample, but >7.5% methylation was prevalent in CGGBP1-depleted sample (values normalized for different number of sequences per sample). I and J: CpG and TpG frequencies on LINE-1 exhibited inverse correlations in CGGBP1 and control shmiR samples establishing the fact that the changes in cytosine content was indeed due to bisulfite conversion of unmethylated cytosines.