Fig. 4From: NCBoost classifies pathogenic non-coding variants in Mendelian diseases through supervised learning on purifying selection signals in humansComparative performance of NCBoost against state-of-the-art methods. The figure shows the area under AUROC (a) and the AUPRC (b) obtained for NCBoost (configuration of features ABCD) together with the six state-of-the-art methods (CADD, DeepSEA, Eigen, Eigen-PC, FunSeq2, and ReMM; “Methods” section) when tested on the same set of positive and negative variants described for Fig. 3Back to article page