Table 1 Performance of the three approaches for Study 1.

From: A model-free and distribution-free multi-omics integration approach for detecting novel lung adenocarcinoma genes

Method

\(\mathcal {S}\)

Selection size

\(\mathcal {P}_{\varvec{X}_1}\)

\(\mathcal {P}_{\varvec{X}_6}\)

\(\mathcal {P}_{\varvec{X}_{12}}\)

\(\mathcal {P}_{\varvec{X}_{22}}\)

\(\mathcal {P}_a\)

SIS

25%

6777.25

\(s_1\)

0

0

0.08

0

0

50%

8118

\(s_2\)

0.01

0.01

0.1

0

0

95%

9887.05

\(s_3\)

0.01

0.01

0.12

0.01

0

DC-SIS

25%

6558.75

\(s_1\)

0

0

0.13

0

0

50%

7935

\(s_2\)

0

0.01

0.2

0

0

95%

9805.25

\(s_3\)

0.01

0.01

0.22

0

0

MrDcGene

25%

28.75

\(s_1\)

0.82

0.84

0.86

0.61

0.31

50%

99.5

\(s_2\)

0.88

0.89

0.93

0.68

0.44

95%

1180.3

\(s_3\)

0.89

0.9

0.95

0.72

0.52

  1. The minimum selection size \(\mathcal {S}\), the individual success rate \(\mathcal {P}_s\), and the overall success rate \(\mathcal {P}_a\) are demonstrated. The predetermined cutoffs for \(\mathcal {P}_s\) and \(\mathcal {P}_a\) are \(s_1 = [n/\log (n)]=37\), \(s_2 = 2s_1\) and \(s_3 = 3s_1\).