S00020: Global Biobank Meta-Analysis
Investigator: Prof Mark Daly
The Global Biobank Meta-analysis Initiative aims to create a framework to jumpstart global collaborations among biobanks. We propose a meta-analysis activity across a number of biobanks with established genotype, phenotype and genomewide association analyses resources in place as one pilot feasibility project towards the development of a more definitive, global resource for human genetics. This is not meant to replace or compete with any existing efforts each of us may be a part of, but rather to jumpstart global efforts by demonstrating the potential of collaboration between groups with mature efforts and analytic expertise to help inform future global activities. The benefits include a power increase for genome-wide association studies (GWASs), the opportunity to rapidly cross-validate genetic findings from individual biobanks, improvements in fine-mapping, and scale to explore more refined subgroups.
Phase 1: Meta-analysis
Overview We here propose a meta-analysis activity across a number of biobanks with established genotype, phenotype and genomewide association analyses resources in place as one pilot feasibility project towards the development of a more definitive, global resource for human genetics. This is not meant to replace or compete with any existing efforts each of us may be a part of, but rather to jumpstart global efforts by demonstrating the potential of collaboration between groups with mature efforts and analytic expertise to help inform future global activities. Even an initial meta-analysis, however, would enable cross-validation of new findings, substantial improvements in fine-mapping as well as more precise estimates of the impact of existing discoveries on broader groups of endpoints and more refined subgroups. Initial pilots, done as demonstrations for the recent ASHG and other meetings, have reported a handful of examples (generally from common endpoints with well-established GWAS such as glaucoma, atrial fibrillation and RA) have already involved FinnGen (FG), Biobank Japan (BBJ), Estonia Biobank (EB), UK Biobank (UKBB), Hunt Biobank (HB), Partners Biobank (PB), Generation Scotland (GS). These initial tests have demonstrated promising concordance and consistency with established GWAS from the respective consortia - suggesting that consistent endpoint definitions for genetic investigation can be defined from data sets and medical systems with different underlying raw data. In these preliminary tests, we implemented FinnGen endpoints defined by clinical expert working teams from academia and industry and then implemented those definitions as closely as possible in the other biobanks. Here we propose a more formal expanded pilot activity to harmonize endpoints for a larger set of 30-40 endpoints and an analytic strategy for a more formal meta-analysis to be delivered publicly via an established web results browser.
