Based on our systematic comparisons of resemblance in DZ twin pairs conditional on sex in a large dataset, we find that sex by genotype interactions for a broad range of behavioral, psychiatric and health related phenotypes are rare. These results suggest that for most variables there is no strong a priori hypothesis that different genes are expressed in men and women. This does not imply that it is not worth to search for sex-specific genes in GWAS because there might still be cases of sex-specific effects at specific loci that are overshadowed by polygenic variance.
