Table 2 Population structure and type I error. No marked differences were observed between the p-values in perfect or imperfect matching scenarios (with or without PCs) to maintain the study-wise type I error rate at 0.05

From: Project MinE: study design and pilot analyses of a large-scale whole-genome sequencing study in amyotrophic lateral sclerosis

Scenario 1: perfect matching

 

Type I error in burden tests

T1

MB

VT

SKAT

No PCs

4.25 × 10−6

3.38 × 10−6

3.94 × 10−6

2.35 × 10−6

Common PCs

2.86 × 10−6

3.99 × 10−6

8.12 × 10−6

2.24 × 10−6

Scenario 2: imperfect matching

 

Type I error in burden tests

T1

MB

VT

SKAT

No PCs

4.40 × 10−6

1.90 × 10−5

1.46 × 10−5

6.72 × 10−6

Common PCs

5.35 × 10−6

9.35 × 10−6

1.04 × 10−9

7.81 × 10−6

Back to article page