Skip to main content

Table 3 Comparison of power between GCDH and single-SNP approaches in analysis of exome-sequencing data from the Rotterdam Study

From: CollapsABEL: an R library for detecting compound heterozygote alleles in genome-wide association studies

 

Causal SNPs available

Causal SNPs excluded

β

MAF

Single-SNP

GCDH

Single-SNP

GCDH

1

(0.00, 0.02)

0.09

0.09

0.08

0.08

1

(0.02, 0.04)

0.17

0.23

0.13

0.14

1

(0.04, 0.06)

0.18

0.28

0.11

0.15

1

(0.06, 0.08)

0.29

0.43

0.19

0.22

1

(0.08, 0.10)

0.20

0.44

0.14

0.18

2

(0.00, 0.02)

0.13

0.16

0.10

0.11

2

(0.02, 0.04)

0.29

0.56

0.19

0.29

2

(0.04, 0.06)

0.28

0.69

0.18

0.32

2

(0.06, 0.08)

0.41

0.75

0.31

0.41

2

(0.08, 0.10)

0.41

0.83

0.28

0.46

3

(0.00, 0.02)

0.18

0.19

0.12

0.12

3

(0.02, 0.04)

0.47

0.72

0.32

0.41

3

(0.04, 0.06)

0.45

0.86

0.32

0.49

3

(0.06, 0.08)

0.55

0.89

0.41

0.58

3

(0.08, 0.10)

0.65

0.94

0.46

0.63

4

(0.00, 0.02)

0.23

0.26

0.13

0.15

4

(0.02, 0.04)

0.55

0.83

0.41

0.53

4

(0.04, 0.06)

0.56

0.94

0.43

0.62

4

(0.06, 0.08)

0.70

0.97

0.54

0.71

4

(0.08, 0.10)

0.75

0.98

0.58

0.76

  1. The simulation analyses are conducted based on the exom sequencing data from Rotterdam Study 1 (RS1), consisting of 1037 individuals and 167,209 SNPs. Power estimates are calculated from 10,000 simulations. Type-I error rate for single-SNP and GCDH analyses are controlled at 5 %