Skip to main content

Table 4 Area under the curve (AUC) calculated from Receiver operating curves for each of the in silico prediction tools alone using their respective probabilities/scores for KCNQ1 , KCNH2 , SCN5A and all genes combined

From: Assessment of the predictive accuracy of five in silico prediction tools, alone or in combination, and two metaservers to classify long QT syndrome gene mutations

Programs

KCNQ1

KCNH2

SCN5A

All genes

AUC

CI

AUC

CI

AUC

CI

AUC

CI

PolyPhen-2

0.942

(0.877-1.000)

0.774

(0.624-0.924)

0.626

(0.485-0.767)

0.769

(0.688-0.850)

SNPs&GO

0.933

(0.882-0.984)

0.864

(0.752-0.976)

0.666

(0.528-0.803)

0.781

(0.714-0.849)

SIFT

0.834

(0.687-0.982)

0.819

(0.729-0.910)

0.643

(0.450-0.836)

0.715

(0.602-0.828)

PROVEAN

0.943

(0.891-0.995)

0.904

(0.840-0.968)

0.631

(0.456-0.807)

0.786

(0.689-0.883)

SNAP

0.689

(0.500-0.877)

0.397

(0.194-0.600)

0.590

(0.421-0.758)

0.627

(0.522-0.732)

Meta-SNP

0.959

(0.918-0.999)

0.905

(0.826-0.984)

0.639

(0.506-0.772)

0.839

(0.781-0.897)

PredictSNP

0.713

(0.571-0.856)

0.549

(0.373-0.725)

0.555

(0.397-0.714)

0.603

(0.513-0.693)