Modeling sex as a personalized biological variable using the Drosophila metabolome

Abstract

The research community now routinely considers “Sex as a Biological Variable” (SABV) in medical research and healthcare, because differences between genders/sexes can significantly impact disease prevalence, progression, and responses to treatment. Personalized medicine considers genetics and SABV, but a truly precise approach requires an understanding of how genetics and sex interact (GxS) to influence an individual’s internal biochemistry. Currently, empirical research is lacking that assesses the degree to which genetic diversity contributes to variability in SABV. Animal models with robust genetic toolkits, such as Drosophila melanogaster, can be used to measure the prevalence of GxS interactions in a population, interactions that are difficult to detect in humans. This dissertation presents the results from two complementary projects measuring GxS interactions in the metabolome of Drosophila–a model for natural genetic variation and a model for monogenic, Mendelian genetic variation in sex characteristics (VSC). This work conceptualizes sex differences in metabolite levels as genotype-specific biological effects. The analyses presented here reveal that the effect of sex on the metabolome is far from binary or dichotomous, but rather falls along a continuum, with genotype playing a larger role than sex alone on most metabolite levels. I show that genetic variation in external, measurable dimorphic traits are associated with genetic variation in sex differences at a molecular level, which effectively renders biological sex a variable as unique and diverse as any individual genome. These studies underscore the importance of considering genetic context in research that incorporates SABV, and the potential pitfalls of analyzing sex as a fixed effect in statistical models.