Fig. 1
From: A theory-based practical solution to correct for sex-differential participation bias
The underlying meaning of sex GWAS. a Observed mean chi-squared statistics over expected chi-squared statistics of sex GWAS with respect to differing participation rates under sex-differential and sex-independent GWAS. Different lines correspond to different liability scale heritabilities of study participation for each sex. b Observed mean chi-squared statistics under sex-differential participation of sex-combined GWAS, sex-stratified GWAS, and fixed-effects meta-analysis of sex-stratified GWAS. c MR-IVW (inverse variance weighted) estimates of two independent binary traits using summary statistics from sex-combined and sex-stratified GWAS. d Difference between marginal effect size of a locus on study participation is the determinant of sex GWAS under study participation. The slope is the difference of study participation rate between male and female. e Genetic correlation of complex traits with biological sex in the UK Biobank. The estimates were adopted from Pirastu et al. [4]