Skip to main content

Table 3 Results of stratified Mendelian randomization analysis

From: Sex and statin-related genetic associations at the PCSK9 gene locus: results of genome-wide association meta-analysis

Subgroup

 

All SNPs

Without rs11583680

 

min (F)

\(\widehat{\theta }\)

\(SE\left(\widehat{\theta }\right)\)

\(p\left(\widehat{\theta }\right)\)

Q

p(Q)

\(\widehat{\theta }\)

Q

p(Q)

Free

77.4

0.94

0.02

 < 1 × 10–200

60.3

5.07 × 10–13

0.98

0.3

0.842

M

71.5

0.92

0.03

 < 1 × 10–200

58.0

1.56 × 10–12

0.98

6.2

0.045

W

41.6

0.92

0.03

 < 1 × 10–200

66.7

2.15 × 10–14

0.99

8.7

0.013

W—free

36.4

0.89

0.03

 < 1 × 10–200

57.7

1.8 × 10–12

0.96

7.3

0.026

M—free

34.7

0.98

0.03

 < 1 × 10–200

17.4

5.95 × 10–4

1.01

2.2

0.331

Treated

29.2

0.66

0.05

1.53 × 10–35

34.8

1.35 × 10–7

0.81

2.1

0.351

M—treated

23.7

0.59

0.06

8.00 × 10–24

26.7

6.78 × 10–6

0.74

5.0

0.082

W—treated

7.1

0.80

0.11

1.66 × 10–13

10.9

1.21 × 10–2

0.91

1.5

0.472

  1. For each subgroup, we report the minimal F-statistics across the four instruments, and the IVW estimates using either all four SNPs (columns 3–6) or leaving rs11583680 out (columns 7–10). Cochran’s Q estimates the heterogeneity of causal estimates across instruments. After removing rs11583680, heterogeneity was markedly reduced. Full results of the other leave-one-out models and for MR-Egger can be found in Additional file 2: Table S8b
Back to article page

Biology of Sex Differences

ISSN: 2042-6410

Contact us