Skip to main content

Table 4 Replication of previously reported heterogeneity by sex in PAGE

From: Investigation of gene-by-sex interactions for lipid traits in diverse populations from the population architecture using genomics and epidemiology study

  

Prior study

PAGE

 

Gene

Trait

SNP

CA

CAF

Beta (SE), male

Beta (SE), female

PAGE SNP

§R2

CA

CAF

Required sample size@for 80% power to detect interaction

Beta (SE), male

Beta (SE), female

Phet

Sex with stronger absolute effect (prior, PAGE)

Significant heterogeneity in PAGE (phet < 0.05)

BUD13

TG

rs11820589

A

0.05

0.131 (0.009)

0.091 (0.008)

rs3135506

0.79

C

0.06

215,337

0.16 (0.02)

0.09 (0.01)

4.3E-4

MM

GCKR

TG

rs1260326

T

0.41

0.064 (0.005)

0.050 (0.004)

rs1260326

-

T

0.42

370,777

0.072 (0.009)

0.046 (0.006)

0.013

MM

LPL

TG

rs13702

T

0.70

0.070 (0.005)

0.047 (0.004)

rs328

0.39

C

0.90

389,318

−0.110 (0.014)

−0.069 (0.009)

0.013

MM

PLTP

HDL

rs6065904

A

0.22

−0.008 (0.003)

−0.031 (0.004)

rs7679

0.86

T

0.82

242,184

0.10 (0.30)

1.68 (0.37)

9.9E-4

FF

TRIB1

TG

rs2954033

A

0.30

0.050 (0.005)

0.022 (0.004)

rs2954029

0.36

A

0.54

137,339

−0.059 (0.010)

−0.031 (0.006)

0.015

MM

No significant heterogeneity in PAGE (phet > 0.05)

APOB

TG

rs11902417

A

0.23

−0.038 (0.006)

−0.024 (0.005)

rs693

0.27

T

0.49

>1,000,000

−0.025 (0.009)

−0.014 (0.007)

0.32

MM

APOB

LDL

rs531819

T

0.15

−0.092 (0.012)

−0.128 (0.012)

rs562338

0.70

T

0.19

126,763

−5.32 (0.73)

−5.96 (0.66)

0.51

FF

CETP

HDL

rs7499892

T

0.17

−0.074 (0.004)

−0.091 (0.005)

rs1864163

0.57

A

0.24

525,658

−2.57 (1.55)

−1.16 (2.68)

0.65

FM

CILP2 ‡

LDL

rs10401969

T

0.07

0.15 (0.02)

0.03 (0.01)

rs16996148

1.0

T

0.08

18,530

−3.44 (0.96)

−2.11 (0.76)

0.28

MM

FADS1

HDL

rs174548

C

0.70

0.015 (0.003)

0.024 (0.004)

rs174547

0.80

T

0.66

986,476

0.82 (0.22)

0.60 (0.20)

0.47

FM

GALNT2

HDL

rs2296065

A

0.85

0.015 (0.004)

0.030 (0.005)

rs2144300

0.26

T

0.60

>1,000,000

0.21 (0.19)

0.63 (0.17)

0.091

FF

GALNT2

TG

rs2296065

A

0.85

−0.015 (0.007)

−0.027 (0.005)

rs2144300

0.26

T

0.60

>1,000,000

0.015 (0.008)

0.019 (0.006)

0.73

FF

LIPC

HDL

rs2070895

A

0.22

0.033 (0.003)

0.051 (0.004)

rs261332

0.91

A

0.20

375,405

1.77 (0.32)

1.72 (0.42)

0.96

FM

LIPG

HDL

rs2156552

A

0.16

−0.023 (0.004)

−0.032 (0.005)

rs2156552

-

T

0.17

>1,000,000

−0.38 (0.24)

−0.56 (0.24)

0.60

FF

LPL

HDL

rs10096633

A

0.40

0.014 (0.032)

0.079 (0.011)

rs328

0.87

G

0.10

53,495

1.76 (0.31)

1.93 (0.28)

0.68

FF

LPL ‡

HDL

rs12678919

A

0.12

−0.20 (0.01)

−0.13 (0.01)

rs328

1.0

G

0.10

40,123

1.76 (0.31)

1.93 (0.28)

0.68

MF

LPL †

HDL

rs2083637

G

0.26

0.149

0.079

rs328

0.43

G

0.10

93,909

1.76 (0.31)

1.93 (0.28)

0.68

MF

PLTP

TG

rs6073952

A

0.19

0.009 (0.006)

0.030 (0.005)

rs7679

1.0

T

0.82

247885

0.013 (0.013)

0.009 (0.011)

0.82

FM

TRIB1

LDL

rs17321515

A

0.52

0.022 (0.008)

0.045 (0.009)

rs2954029

0.97

A

0.54

63691

1.98 (0.67)

2.16 (0.49)

0.83

FF

  1. *Abbreviations: CA = coding allele; CAF = coding allele frequency; SE = standard error; Phet = heterogeneity by sex p-value.
  2. All prior SNPs shown are from Asselbergs et al. (2012), except for the three noted.
  3. †Aulchenko, Ripatti et al., 2009. Standard errors were not reported for sex-specific effects.
  4. ‡Teslovich, Musunuru, et al., 2010. Effect sizes were reported for standardized residuals and therefore are not directly comparable to the PAGE results.
  5. §LD was calculated using 1000 Genomes Pilot 1 data (CEU population).
  6. %Criteria for replication: phet < 0.05 in PAGE and consistent direction of interaction effect (e.g., stronger effects in females in both PAGE and prior study).
  7. @Sample size calculations for gene-environment interactions done on Quanto, assuming the CAF from PAGE and the difference in βf-βm from the prior study. Calculated sample sizes were then divided by R2 (the LD between the PAGE SNP and the prior SNP) to arrive at the final required sample size for 80% power, shown in the table.