A large-scale population study has uncovered significant differences between the metabolite profiles of men and women, and it has identified SNPs that are associated with the sexual dimorphism of one particular metabolite. This work underscores the importance of stratifying by sex in trait-mapping studies.

In this systematic survey, fasting serum concentrations of 131 metabolites were examined in >3,300 individuals from KORA F4, a German population cohort. Linear regression analysis showed that sex accounted for differences between men and women in the concentration of each of 102 of these metabolites, an effect that was replicated in a smaller population (KORA F3). In each case, sex specificity was not affected by adjusting the data for covariates such as smoking or type 2 diabetes.

Correlation patterns in the metabolic network were broadly consistent with the subdivision of the metabolites into five chemically distinct subgroups. For example, metabolites from the same subgroup were more likely to be higher in one sex than in the other.

To locate the genomic determinants of the observed metabolic differences, a sex-stratified genome-wide association study was carried out on a portion of the KORA F4 cohort. This investigation, which was partly replicated in subsamples of KORA F4 and KORA F3, identified eight SNPs at the CPS1 locus that were associated with differences between men and women in glycine levels.

The correlation between sex and disease incidence, severity and age of onset are well documented in many species; the strength of this study lies in exploring the phenomenon of gender differences on such a large scale.