Skip to main content

Table 2 Power of GENN to detect correct functional epistatic loci.

From: Power of grammatical evolution neural networks to detect gene-gene interactions in the presence of error

Source of Noise

Power (%)

 

Model 1

Model 2

Model 3

Model 4

Model 5

Model 6

None

100

100

96

91

69

72

GE

100

100

96

85

58

68

GH

7

4

15

16

14

16

PC

88

92

21

12

17

21

MS

100

100

99

82

42

74

GE+GH

3

6

14

16

11

9

GE+PC

92

95

19

11

12

16

GE+MS

100

100

93

75

48

58

GH+PC

9

9

13

15

10

11

GH+MS

1

0

0

0

0

1

PC+MS

65

85

18

13

7

9

GE+GH+PC

3

9

2

2

7

3

GE+GH+MS

5

1

0

0

0

0

GH+PC+MS

0

0

0

0

0

0

GE+PC+MS

62

81

14

9

9

9

GE+GH+PC+MS

0

0

0

0

0

0

  1. GE = 5% Genotyping Error; GH = 50% Genetic Heterogeneity; PC = 50% Phenocopy; MS = 5% Missing Data