10.6084/m9.figshare.10027106.v1 Reka Toth Reka Toth Heiko Schiffmann Heiko Schiffmann Claudia Hube-Magg Claudia Hube-Magg Franziska Büscheck Franziska Büscheck Doris Höflmayer Doris Höflmayer Sören Weidemann Sören Weidemann Patrick Lebok Patrick Lebok Christoph Fraune Christoph Fraune Sarah Minner Sarah Minner Thorsten Schlomm Thorsten Schlomm Guido Sauter Guido Sauter Christoph Plass Christoph Plass Yassen Assenov Yassen Assenov Ronald Simon Ronald Simon Jan Meiners Jan Meiners Clarissa Gerhäuser Clarissa Gerhäuser MOESM5 of Random forest-based modelling to detect biomarkers for prostate cancer progression Springer Nature 2019 differentially methylated CpG sites TCGA BCR prostate cancer progression CpG sites Random forest-based modelling PRAD ICGC prostate cancer methylation beta values MOESM 2019-10-23 08:24:23 Journal contribution https://springernature.figshare.com/articles/journal_contribution/MOESM5_of_Random_forest-based_modelling_to_detect_biomarkers_for_prostate_cancer_progression/10027106 Additional file 5: Figure S3. Heatmap of the selected CpG sites in the ICGC prostate cancer (left) and TCGA PRAD (right) validation datasets. Each column represents a sample with predicted good or poor prognosis, while rows represent selected differentially methylated CpG sites. Annotations on the left side indicate top ranked candidate genes associated with most informative CpG sites. Low and high methylation beta values in a range from 0 to 1 are shown in a blue to red color scale. BCR: PSA-based biochemical recurrence.