Table 5. MLP’s and sdAE’s ROC curve AUC with 95% confidence intervals evaluated on subsets of SNVs from the training dataset II that could not be processed by other predictors.

Score	Missing predictions	MLP	sdAE
Eigen	1175	0.97 (0.94–0.98)	0.95 (0.92–0.97)
FATHMM	898	0.80 (0.75–0.85)	0.34 (0.28–0.42)
LRT	1772	0.94 (0.93–0.95)	0.80 (0.77–0.84)
MetaLR	118	0.76 (0.68–0.85)	0.59 (0.49–0.70)
MetaSVM	118	0.76 (0.67–0.85)	0.59 (0.48–0.69)
MutationAssessor	843	0.90 (0.88–0.92)	0.72 (0.67–0.77)
PROVEAN	426	0.85 (0.81–0.90)	0.47 (0.39–0.55)
PolyPhen HDIV	286	0.85 (0.80–0.89)	0.53 (0.46–0.60)
PolyPhen HVAR	286	0.84 (0.80–0.89)	0.53 (0.45–0.61)
SIFT	514	0.89 (0.85–0.92)	0.59 (0.52–0.66)