Table 2 Summary phenotypic variance explained by SNPs used in MR analyses for BMI

From: Effect of increased body mass index on risk of diagnosis or death from cancer

Trait

Discovery sample size

Relevant UKBB fields

Number of SNPs

r 2a

Description

BMI

390,628

UKB Field ID 50

520

0.07

SNP association adjusted for age and sex.

Filtered BMI

390,628

UKB Field ID 50

377

0.04

A subset of the BMI SNPs that are not associated (with P > 1e–5) to potential confounders (i.e., smoking phenotypes, alcohol intake, coffee/tea, height). For sensitivity analyses.

  1. SNPs associated with the trait at P   < 1e–8 were used as instruments. SNPs were pruned for LD at r2  < 0.01 to ensure strict independence between instruments
  2. UKBB UK Biobank, BMI body mass index, MR Mendelian randomization
  3. ar2; proportion of phenotypic variance explained by SNPs. The phenotypic variance of trait Y explained by SNPs were calculated based on \(\mathop {\sum }\limits_i 2p_i\left( {1 - p_i} \right)\beta _i^2/Var(Y)\), where pi and βi refers to the minor allele frequency and the magnitude of association of the i-th SNP instrument on trait Y
Back to article page