Fig. 3From: TADA—a machine learning tool for functional annotation-based prioritisation of pathogenic CNVsPredictive Performance of the Deletion and Duplication Classifiers. A and B show the ranking performance of TADA and SVScore for deletions and duplications, respectively. For each bin we computed the percentage of variants placed amongst the corresponding rank or ranks. Black bars indicate the standard variation based on 30 random sampling runs. C shows the ROC-Curves and AUC values for the deletion and duplication classifiers based on the Test-Split and ClinVar variants for both TADA and SVScoreBack to article page