Table. 2 Linear relationships between observed traits and PS in UK Biobank

From: Apparent latent structure within the UK Biobank sample has implications for epidemiological analysis

Observed trait (unit)

Model 1

Model 2

Model 3

Model 4

Model 1

Model 2

Model 3

Model 4

 

PS for BMI (GIANT)

Household income (£ per year)

−335 (1.8e−9)

−325 (5.2e−9)

−251 (4.0e−6)

−229 (3.4e−5)

−304 (4.7e−8)

−294 (1.3e−7)

−212 (1.0e−4)

−190 (0.0057)

BMI (kg/m2)

0.612 (<2e−16)

0.611 (<2e−16)

0.606 (<2e−16)

0.606 (<2e−16)

0.549 (<2e−16)

0.547 (<2e−16)

0.541 (<2e−16)

0.541 (<2e−16)

Age at completion of full-time education (years)

−0.0219 (3.2e−4)

−0.0216 (4.0e−4)

−0.0201 (9.2e−4)

−0.0187 (0.0025)

−0.0231 (1.6e−4)

−0.0227 (2.0e−4)

−0.0201 (9.6e−4)

−0.0187 (0.0024)

Number of siblings (persons)

0.0107 (3.0e−4)

0.0105 (3.6e−4)

0.00783 (0.0071)

0.00750 (0.011)

0.00130 (1.0e−05)

0.00129 (1.3e−05)

0.00850 (0.0035)

0.00807 (0.0068)

 

PS for EA (SSGAC)

Household income (£ per year)

1066 (<2e−16)

1062 (<2e−16)

874 (<2e−16)

835 (<2e−16)

1454 (<2e−16)

1446 (<2e−16)

1200 (<2e−16)

1140 (<2e−16)

BMI (kg/m2)

−0.112 (<2e−16)

−0.111 (<2e−16)

−0.101 (<2e−16)

−0.097 (<2e−16)

−0.151 (<2e−16)

−0.150 (<2e−16)

−0.132 (<2e−16)

−0.129 (<2e−16)

Age at completion of full-time education (years)

0.0878 (<2e−16)

0.0877 (<2e−16)

0.0844 (<2e−16)

0.0831 (<2e−16)

0.12 (<2e−16)

0.119 (<2e−16)

0.112 (<2e−16)

0.109 (<2e−16)

Number of siblings (persons)

−0.0250 (<2e−16)

−0.0250 (<2e−16)

−0.0258 (<2e−16)

−0.0253 (<2e−16)

−0.038 (<2e−16)

−0.0382 (<2e−16)

−0.0293 (<2e−16)

−0.0279 (<2e−16)

 

PS for height (GIANT)

Household income (£ per year)

522 (<2e−16)

515 (<2e−16)

418 (1.8e−14)

406 (2.7e−13)

515 (<2e−16)

509 (<2e−16)

419 (1.7e−14)

405 (2.9e−13)

BMI (kg/m2)

−0.129 (<2e−16)

−0.128 (<2e−16)

−0.112 (<2e−16)

−0.116 (<2e−16)

−0.122 (<2e−16)

−0.121 (<2e−16)

−0.105 (<2e−16)

−0.109 (<2e−16)

Age at completion of full-time education (years)

0.0350 (9.4e−9)

0.0348 (1.1e−8)

0.0289 (2.0e−06)

0.0263 (2.0e−05)

0.0349 (1.1e−08)

0.0347 (1.2e−08)

0.0286 (2.6e−6)

0.0265 (1.8e−5)

Number of siblings (persons)

−0.0249 (<2e−16)

−0.0248 (<2e−16)

−0.0130 (8.1e−06)

−0.0119 (7.2e−05)

−0.0264 (<2e−16)

−0.0263 (<2e−16)

−0.0136 (3.0e−6)

−0.0127 (2.1e−5)

 

Weighted PS (p < 5e−8))

Unweighted PS (p < 5e-8)

  1. The field contents are beta coefficients per 1 SD increase in PS, with p values for the linear association, testing the null hypothesis of no linear association between each observed trait and PS in brackets. For household income, N = 276,779; BMI, N = 336,031; age at completion of full-time education, N = 228,886; number of siblings, N = 332,037. Statistical adjustment was performed as follows: model 1: no adjustment; model 2: adjustment for genotyping array only; model 3: adjustment for genotyping array, 40 PCs and study participation centre; model 4: adjustment for genotyping array, 40 PCs, study participation centre and non-linear regression terms for North and East axes of birth location
  2. PS polygenic score, PC  principal component, BMI  body mass index, EA  educational attainment, GIANT Genetic Investigation of ANthropometric Traits, SSGAC Social Science Genetic Association Consortium
Back to article page