Table 3 Optimal cut-off value for each individual CpG

From: A novel analytical model of MGMT methylation pyrosequencing offers improved predictive performance in patients with gliomas

CpG sites

Cut-off value

AUC

Standard error

P

95% confidence interval

Sensitivity

Specificity

Lower

Upper

75

16

0.693

0.060

0.002

0.576

0.809

71%

68%

78

4

0.682

0.060

0.004

0.566

0.799

76%

60%

81

11

0.678

0.060

0.006

0.560

0.797

68%

68%

82

15

0.674

0.061

0.007

0.554

0.793

65%

70%

76

12

0.673

0.060

0.007

0.555

0.791

71%

64%

77

10

0.659

0.062

0.013

0.538

0.781

56%

76%

80

15

0.659

0.062

0.014

0.538

0.780

62%

70%

79

5

0.644

0.062

0.026

0.522

0.765

65%

64%

  1. AUC area under the curve of the receiver operating characteristic curve analysis