The Architecture of Genome Wide Association Studies

Over 3,000 genome-wide association studies (GWAS) across a decade of research have discovered genetic risk variants and biological pathways that play a role in specific diseases, disorders and behavioural traits. The NHGRI-EBI GWAS Catalog indexes all published studies, to which we construct, harmonise and then merge various data sources and conduct multiple detailed analyses in order to provide the first systematic, data-driven examination of GWAS to date. We pursue multiple distinct avenues of interest; participants, funders, researchers, traits, cohorts and consortium. We first replicate and substantially expand existing analysis with regard to participant ancestry over time, adding further evidence of bias towards recruitment from European and American countries. We then provide an analysis of who funds GWAS (and what they fund): over 85% of acknowledgements relate to NIH grants, with almost the entirety of the remainder relating to UK agencies. A network analysis of GWAS authors shows a dense group of Principle Investigators contributing large datasets, reflected in centrality measures and an original ‘GWAS H-Index’. Approximately 40% of authors are female, although men dominate the typical ‘senior author’ position (71%). We also find a gendered division across the studied traits, where the dominant categories relate to neurological disorders, neoplasm and drug responses. An extensive manual data collection exercise summarises the most frequently utilised cohorts for which we collate a range of summary descriptives, and we also provide an overview of the main consortium in play. In light of these findings, we discuss challenges and solutions for the future of genomics.