Skip to main content
. 2015 Jan 5;22(2):131–144. doi: 10.1530/ERC-14-0454

Figure 2.

Figure 2

Validation of HES5 promoter methylation as a common event in two additional independent prostate cancer cohorts. (A) CpG methylation summary of the HES5 promoter as determined by bisulphite sequencing from a representative tumour–normal pair. Each column represents one CpG assayed (n=60), red and blue stacked bars represent the proportion of methylated and unmethylated reads, respectively, at each CpG. Column widths are proportional to sequencing coverage (median=786×). (B) Scatter plot summary of HES5 promoter methylation for 39 tumour–normal pairs. (C) Histogram summary of significance testing for increased HES5 promoter methylation in tumour vs normal sample pairs (n=39 pairs from panel-C; paired Wilcox rank sum test; −log2 P values are plotted to visualise distributions). (D) Boxplot summary of HES5 promoter methylation for 304 tumour and 49 benign prostate samples on Illumina 450k arrays (TCGA data). (E) Histogram summary of significance testing for increased HES5 promoter methylation in TCGA tumour vs normal sample pairs (n=49 pairs from panel-E; paired Wilcox rank sum test; −log2 P values are plotted to visualise distributions). (F) ROC curve for HES5 promoter methylation using data from bisulphite sequencing of 39 tumour normal pairs (A, B and C) and methylation array profiling of 49 tumour normal pairs (D and E).