Prioritizing diversity in polygenic risk prediction of primary open-angle glaucoma

点击翻译按钮获取中文摘要

PROJECT SUMMARY: Prioritizing diversity in polygenic risk prediction of primary open-angle glaucoma Glaucoma is the leading worldwide cause of irreversible blindness. Primary open-angle glaucoma (POAG), the most common type of glaucoma, is more prevalent and severe in individuals of African ancestry. Unfortunately, individuals from this ancestral group have been under-represented in genome-wide association studies (GWAS) thus far. Furthermore, polygenic risk scores (PRS) based on GWAS data from European-descent populations are not transferable to individuals of diverse (non-European) ancestry. Given the aspirations of precision medicine, PRS demonstrate clinical potential but fall short, in part, due to the lack of diversity in these studies. We hypothesize that clinically-implementable PRS can be achieved by including African and African- descent individuals in gene discovery and PRS development. To inform and improve precision ocular health, we will prioritize diversity in polygenic risk prediction of POAG with three proposed aims that will yield the largest-ever meta analyses of POAG in African and African-descent individuals and the first-ever African-ancestry focused POAG PRS. In Aim 1, we will perform meta-analyses of 47,078 samples (18,037 cases, 29,041 controls) to identify novel POAG loci in African and African-descent populations. Given that most GWAS have been performed in mainly European and European-descent populations, we hypothesize that undiscovered POAG risk loci will be detected by leveraging the power of meta-analysis of case-control GWAS in African and African-descent population samples. In Aim 2, we will meta-analyze admixture mapping results for each of our datasets to identify African ancestry-specific POAG loci. We hypothesize that admixture mapping will identify genomic regions where African ancestry co- segregates with POAG risk. Significant loci from Aims 1 and 2 will be fine-mapped and evaluated for selection signatures. In Aim 3, we will build and optimize POAG PRS in African and African-descent base dataset meta- analyses. We will then evaluate novel and published PRS for POAG classification in test datasets. We hypothesize that ancestrally-informed POAG PRS will better predict POAG and relevant clinical outcomes in African and African-descent populations compared to those derived from primarily European-descent data. Meta-analysis results from Aims 1 and 2 as well as variants from our optimized PRS in Aim 3 will undergo pathway analyses to identify biological pathways and statistical driver genes implicated in POAG risk. In the long-term, we hope that the information gained from this project will inform a broader understanding of POAG genetics across diverse ancestry groups and provide the foundational basis for the clinical applicability of ancestrally-informed PRS for POAG.