Table 1

MCC used to evaluate the performance of the seven tools

GenesSIFTPolyPhen2VEST4REVELReVeClinPredCAPICE
G6PD0.410.420.590.610.560.520.33
ALAS20.410.630.610.730.58
RS10.380.430.690.590.690.690.43
MTM10.600.670.710.580.560.660.55
OTC0.380.380.610.460.460.650.51
PHEX0.420.550.620.580.480.620.41
F80.380.610.750.820.770.740.63
IL2RG0.520.590.740.780.610.780.58
L1CAM0.420.490.730.650.640.580.51
CLCN50.550.650.590.420.560.580.53
IDS0.620.720.680.640.700.790.67
GLA0.260.290.570.530.510.520.31
ABCD10.600.630.720.680.690.750.53
F90.250.300.390.570.590.500.41
GJB10.310.350.490.670.530.530.34
AVPR20.560.590.770.620.750.730.65
PDHA10.620.530.700.600.550.800.57
BTK0.690.520.800.800.760.720.58
OCRL0.540.760.780.760.680.620.71
NDP0.570.640.590.750.48
HPRT10.440.240.730.620.650.790.48
  • *As suggested by their respective authors, the pathogenicity thresholds used for SIFT, PolyPhen2, VEST4, REVEL, ReVe, ClinPred and CAPICE were 0.05, 0.85, 0.5, 0.5, 0.7, 0.5 and 0.02, respectively.

  • †The highest MCC value for each gene is highlighted in bold.

  • ‡The SIFT and VEST4 prediction scores for ALAS2 and NDP variants in the transcript of interest were unavailable.

  • MCC, Matthews correlation coefficient.