Table 3 Diagnostic MiRs identification.

From: Identification of novel diagnostic and prognostic microRNAs in sarcoma on TCGA dataset: bioinformatics and machine learning approach

Marker(s)

AUC

SE

SP

Cut-off

NPV

PPV

GLM analysis

Coefficients

Degree of Freedom

AIC

Null Deviance

Residual Deviance

miR.3688

miR.3936

0.76

0.90

0.50

0.98

0.03

0.99

23.45

-2.29

258

26.92

23.44

20.92

miR.1255a

miR.1292

miR.3688

0.90

0.86

1.00

0.97

0.05

1.00

25.44

-0.40

25.83

258

25.44

23.44

17.44

miR.1255a

miR.1292

miR.3678

0.86

0.83

1.00

0.97

0.04

1.00

25.50

-0.14

0.03

258

26.48

23.44

18.48

miR.1255a

miR.1292

0.86

0.82

1.00

0.97

0.04

1.00

25.49

-0.13

258

24.48

23.44

18.48

miR.1255a

miR.1292

miR.3936

0.90

0.78

1.00

0.98

0.03

1.00

25.34

0.22

-1.94

258

25.47

23.44

17.47

miR.1255a

miR.3678

miR.3688

0.89

0.77

1.00

0.98

0.03

1.00

25.34

-0.15

25.75

258

25.52

23.44

17.52

miR.1255a

0.85

0.71

1.00

0.98

0.02

1.00

25.49

258

22.50

23.44

18.5

  1. AIC Akaike information criterion, AUC area under curve, NPV negative predictive value, PPV positive predictive value, SE sensitivity, SP specificity.