Fig. 3

Polygenic methylation score (PMS) for detection of PCa. a Pearson correlation analysis of a weighted PMS of three target CGs (DMPs) in T cell DNA and Gleason score. DMPs were identified using ChAMP package in R. These three DMPs were combined to calculate a PMS for each sample which was plotted against Gleason score using a multivariate linear regression model as described in the Results section. b Heatmap of the T cell DNA methylation levels of the 3 DMPs for all negative-biopsy controls and positive-biopsy PCa patients (High methylation Red; Low methylation, Blue). c ROC curve showing the sensitivity and specificity of PMS score. d Mean percentage tumor volume (%) of control (Ctrl), LGPCa (Low Gleason PCa) and HGPCa (High Gleason PCa) samples predicted from core prostate biopsies as described in Methods. Ctrl (Control, Blue); LGPCa (Red); and HGPCa (Green). e Linear regression analysis of the PMS from T cell DNA with percentage tumor volume. The red dotted lines show the 95% confidence band of the line of best fit. The negative-biopsy control samples were assumed to have a tumor volume of 0% and a Gleason score of 0