A novel approach for equitable characterization of gender and its use in exposing subgroup discrepancies in polygenic score associations

点击翻译按钮获取中文摘要

ABSTRACT Polygenic scores — summaries of the genomic contribution to risk and resilience for biomedical traits — are an emerging and promising approach for clinical risk assessment and personalized medicine. Minoritized groups, such as gender minorities, do not currently benefit from insights gained via studies of polygenic scores because these groups have not been sufficiently included or characterized in this research. This disparity must be addressed both by inclusion during recruitment as well as consideration of the effects of heterogeneity during analysis. For example, our preliminary data show striking dissociations in suicide risk as influenced by polygenic scores. Notably, these opposing associations are apparent only when gender diversity status (i.e., cisgender; gender-diverse) is modeled, underscoring the imperative to parsing the heterogeneity of such associations by both sex and gender. These results suggest value in genomic research for gender minority groups to understand both innate risk and resilience. The promise of genomic research informed by designated sex, gender, and their interaction (i.e., gender diversity) depends on the ability to rigorously and equitably include and characterize individuals across the spectra of designated sex and gender. Currently, biomedical research relies on checklist or write-in gender identity descriptors, which do not capture the continuous and simultaneous nature of dimensional binary and nonbinary gender experiences and result in statistically underpowered analytics with far too few individuals within each gender self-descriptor category. This perpetuates the exclusion of gender and its intersection with designated sex in genetic research. We propose to close this gap by calibrating and genetically characterizing the Gender Self-Report (GSR), a novel and broadly disseminable method for obtaining multidimensional gender for genomic research and broader research applications. First, we will facilitate partnership between scientists and gender diverse community stakeholders, and reduce paternalistic tendencies in this ethically complex field of research, by building on our established community partnerships with a purposively recruited stakeholder panel (N=50) to provide a final version of the GSR itemset. Next, to advance the dimensional characterization of gender identity and gender diversity in a genomic research context, we will validate the stakeholder-refined GSR in a large sample (N=10,000) of genotyped neurotypical and neurodivergent adults, enriched across broad experiences of gender diversity. Finally, to demonstrate proof-of-principle for how designated sex, gender, and gender diversity contextualize patterns of association with polygenic scores, we will measure key health outcomes (both mental and physical), in a large, genetically informed sample enriched for neurodiversity (e.g., autism) and broad gender diversity. The proposed research will provide value to gender minority groups by seeking a better understanding of how polygenic scores apply specifically to them, and not just the cisgender proportional majority.