Skip to main content

Table 2 Performance on 66 HMBS variants with clear clinical annotations taken from ClinVar [44], using all predictors assessed as well as experimental fitness data obtained by yeast complementation assays [44]. The best score for each metric is indicated in bold

From: FiTMuSiC: leveraging structural and (co)evolutionary data for protein fitness prediction

Method

Sensitivity

Specificity

BACC

AUC-ROC

Experimental

0.81

0.92

0.87

0.92

FiTMuSiC

0.96

0.92

0.94

0.98

FATHMM

1.00

0.00

0.50

0.79

PROVEAN

0.96

0.77

0.87

0.87

DEOGEN2

1.00

0.23

0.62

0.93

PolyPhen-2

0.98

0.31

0.64

0.91

EVE

0.94

0.92

0.93

0.98

Sequence UNET

0.70

0.77

0.73

0.82

MutPred2

1.00

0.54

0.77

0.96

AlphaMissense

0.98

0.92

0.95

0.99