Table 3 Performance of the computational tools assessed in this study (in tasks involving PAX6 missense variant evaluation): secondary analysis.
From: Using computational approaches to enhance the interpretation of missense variants in the PAX6 gene
Tool | Threshold | Sp (%) | Sn (%) | Acc (%) | PPV (%) | MCC |
|---|---|---|---|---|---|---|
AlphaMissense | Default thresholds | 65 | 100 | 69 | 30 | 0.44 |
BayesDel | 6 | 100 | 19 | 14 | 0.09 | |
CADD | 3 | 100 | 16 | 14 | 0.06 | |
ClinPred | 12 | 100 | 24 | 15 | 0.14 | |
Eigen | 15 | 100 | 26 | 14 | 0.14 | |
MutPred2 | 34 | 100 | 43 | 19 | 0.25 | |
PolyPhen2 | 65 | 100 | 69 | 30 | 0.44 | |
REVEL | 37 | 100 | 45 | 20 | 0.27 | |
SIFT4G | 79 | 90 | 80 | 39 | 0.50 | |
VEST4 | 20 | 100 | 31 | 16 | 0.18 | |
AlphaMissense | Optimized thresholds | 88 | 90 | 88 | 53 | 0.63 |
REVEL | 82 | 100 | 84 | 46 | 0.61 | |
SIFT4G | 83 | 90 | 84 | 45 | 0.56 | |
Combination (AlphaMissense + SIFT4G + REVEL) | 85 | 90 | 85 | 47 | 0.58 |
