Skip to main content

Table 3 FuNeL success rate in identification of disease-predicting SNPs

From: Functional networks inference from rule-based machine learning models

 

5 SNP

10 SNP

15 SNP

20 SNP

25 SNP

Her.

L-EDM

H-EDM

L-EDM

H-EDM

L-EDM

H-EDM

L-EDM

H-EDM

L-EDM

H-EDM

0.001

6 %

16 %

8 %

18 %

4 %

10 %

4 %

12 %

12 %

16 %

0.005

8 %

82 %

0 %

86 %

6 %

80 %

2 %

82 %

8 %

72 %

0.01

8 %

96 %

8 %

100 %

8 %

100 %

12 %

100 %

14 %

100 %

0.05

14 %

100 %

60 %

100 %

42 %

100 %

34 %

100 %

34 %

100 %

0.1

100 %

100 %

100 %

100 %

100 %

100 %

100 %

100 %

100 %

100 %

0.2

100 %

100 %

100 %

100 %

100 %

100 %

100 %

100 %

100 %

100 %

0.3

100 %

100 %

100 %

100 %

100 %

100 %

100 %

100 %

100 %

100 %

0.4

100 %

100 %

100 %

100 %

100 %

100 %

100 %

100 %

100 %

100 %

  1. The datasets differed with respect to heritability, number of SNPs and detection difficulty (L-EDM models were the hardest, H-EDM the easiest)