Abstract
We examined sex differences in the genetics of thirty-three blood and urine lab test levels (biomarkers) in the UK Biobank population. We built a Bayesian Mixture Model to estimate the fraction of genetic variance shared between men and women for a given trait and used an extension of this model to identify genetic variants with sex-specific effects. The majority of these biomarkers did not show sex-related effects, with the exception of testosterone. For testosterone, we found a large number of genetic variants that showed effects in males but not females and vice versa. We built models to predict testosterone level from genetics, and found that sex-specific models outperformed sex combined models. Additionally, we used the sex-specific genetic variants associated with testosterone to examine the relationship between testosterone and several phenotypic and disease traits. We found evidence for a relationship between testosterone and height, body mass index, waist and hip circumference, and type 2 diabetes. Overall, our results demonstrate that while sex has a limited role in most of the genetics of most of the lab tests we examined, it plays a large role in the genetics of testosterone.
1 Application
Application ID | Title |
24983 | Generating effective therapeutic hypotheses from genomic and hospital linkage data |
1 Return
Return ID | App ID | Description | Archive Date |
3662 | 24983 | Sex-specific genetic effects across biomarkers | 22 Jul 2021 |