Some observed genetic correlations between human traits can be explained by cross-trait assortative mating – an individual’s tendency to choose mates with specific phenotypic characteristics that have no genetic relationship – rather than through widespread pleiotropic effects, according to a new study. The findings suggest that previous studies likely overestimate the true genetic similarity between many phenotypes, including psychiatric disorders. Genome-wide association studies (GWASs) have identified genetic variants correlated with specific traits. However, most traits are correlated with thousands of variants, and many variants are pleiotropic, meaning they are associated with multiple traits. For example, psychiatric traits, including disorders like depression or anxiety, likely share thousands of overlapping genetic variants with one another and even with other conditions across disease categories. This observation of genetic correlations between disparate traits has been used as evidence of widespread pleiotropy across many human phenotypes. Here, Richard Border and colleagues evaluate an overlooked potential source of bias in these findings – cross-trait assortative mating (xAM). To test this hypothesis, Border et al. developed a large atlas of cross-mate correlations across a broad array of phenotypes using two large population-based samples encompassing more than 800,000 individuals and combined it with computer simulations. The authors found that many genetic correlations between human traits can be explained by xAM and that cross-mate phenotype correlations among many pairs of phenotypes are strong enough that one or few generations of xAM would significantly inflate genetic correlation estimates. “The findings of Border et al. make it clear that more realistic models for why mates correlate within and between multiple traits need to be developed and tested,” write Andrew Grotzinger and Matthew Keller in a related Perspective.
Cross-trait assortative mating is widespread and inflates genetic correlation estimates
Article Publication Date