Fig. 4
From: Prenatal BRCA1 epimutations contribute significantly to triple-negative breast cancer development
Similar properties of BRCA1 methylation in blood samples of healthy newborn girls and adult breast cancer patients. A Overall model for early prenatal (constitutional) BRCA1 methylation as an underlying contributor to TNBC. Red dots represent BRCA1-methylated normal cells, appearing through an early embryonic event, resulting in a mosaic adult. Red star represents breast cancer. B Allele specificity of BRCA1 methylation in blood samples from newborn girls heterozygous for SNP rs799905 (N = 40). Preferential methylation of one of the alleles (fold enrichment) is indicated on the x-axis and degree of methylation (VEF value for region CpG14–34) is indicated on the y-axis. Data points in the right half of the plot indicate methylated alleles to be predominantly rs799905 reference alleles, while data points towards the left indicate methylated alleles to be predominantly rs799905 alternative alleles. Gray area above the plot shows smooth kernel density estimates for fold enrichment values. Inf, infinity value, i.e., exclusive methylation of a single allele. C Smoothed averaged CpG methylation levels (y-axis) within assayed genomic region (x-axis) in blood of BRCA1 methylation-positive newborn girls (N = 113; green lines) and breast cancer patients (N = 44; red lines). Solid lines represent averages for all hypermethylated epialleles (per-epiallele average beta value ≥ 0.5); dashed lines represent averages for all hypomethylated epialleles (per-epiallele average beta value < 0.5); light gray areas represent 95% CI. Bars on top represent amplicons, with the bright green ones covering CpGs 14–34. Arrows show BRCA1 and NBR2 transcription start sites; vertical dotted line marks position of SNP rs799905 (see Supplementary Information for more details). D Average beta values (y-axis) of ranked epialleles (x-axis) in blood samples of BRCA1 methylation-positive newborn girls (N = 113; top) and breast cancer patients (N = 44; bottom). All epialleles of the region CpG14–34, within each sample, were ranked by increasing average beta value with every rank centered at epiallele with average beta value of 0.5. Lines connect increasing beta values and represent individual samples. Maximum 5000 epialleles are plotted per sample (beta = 0.5, ± 2500 alleles). The sharp incline in average beta value around beta = 0.5 reveals that most alleles are either hypomethylated or hypermethylated; very few alleles have intermediate methylation levels