Skip to main content

Table 2 Performances on agree and not agree subset of predictions

From: Improving the prediction of disease-related variants using protein three-dimensional structure

 

Q2

P[D]

S[D]

P[N]

S[D]

C

AUC

PM

SEQ∩3D

0.87

0.85

0.89

0.89

0.84

0.74

0.93

91

SEQ-3D

0.66

0.70

0.70

0.62

0.62

0.32

0.71

9

3D-SEQ

0.34

0.38

0.30

0.30

0.38

-0.32

0.35

9

  1. SEQ∩3D indicates the subset of agree predictions between SVM-3D and SVM-SEQ. SEQ-3D and 3D-SEQ are respectively the predictions of SVM-SEQ and SVM-3D on the not agree prediction subset. The accuracy measures are defined in Methods section. PM is the fraction of the dataset. D and N are disease-related and neutral variants respectively.