Enter Note Done


Decrease font size Increase font size

PgmNr 58: Global Biobank Meta-analysis Initiative: Powering genetic discovery across human diseases.

W. Zhou 1; B.M. Neale 1; M.J. Daly 1,2; on behalf of Global Biobank Meta-analysis Initiative

View Session  Add to Schedule

1) Broad Institute, Massachusetts General Hospital, Harvard Medical school, Boston, Massachusetts.; 2) Institute for Molecular Medicine Finland, HiLIFE, University of Helsinki, Helsinki, Finland

With electronic health records and questionnaires linked to genomic data, biobanks provide unprecedented opportunities for systematically understanding genetic and environmental contributions to complex diseases. The Global Biobank Meta-analysis Initiative aims to create a framework to jumpstart global biobank collaboration. The benefits include better power for genome-wide association studies (GWASs), GWASs of understudied diseases, the opportunity for cross-validation, improvements in fine-mapping, and potential to explore subgroup analyses. With scale comes the opportunity to develop collaborations around rare or understudied traits as well as deepen genetic analysis of longitudinal phenotypes and trajectories. 18 biobanks have committed to this project so far, including Biobank Japan, BioME (USA), BioVu (USA), China Kadoorie, Colorado Biobank (USA), deCODE Genetics (Iceland), East London Genes & Health (UK), Estonian Biobank, FinnGen (Finland), Generation Scotland, HUNT (Norway), LifeLines (Netherlands), Mexico City, Michigan Genomics Initiative (USA), Million Veteran Program (USA), Partners Biobank (USA), UCLA Precision Health Biobank (USA) and UK Biobank, bringing the total sample size to more than 2 million.

We describe here initial pilot uses of this collaborative network. We harmonized phenotype definition and performed GWAS across biobanks for pilot common endpoints: asthma, atrial fibrillation (AFib), rheumatoid arthritis (RA), glaucoma, and colorectal cancer. Meta-analysis boosted the number of genome-wide significant loci compared to the sum of loci from individual biobanks. For example, 103 loci were identified for AFib with a total of 46,176 cases across UK Biobank (UKB), FinnGen, HUNT, Partners, Generation Scotland, and Biobank Japan while 50 loci were identified from the union of individual biobanks’ results and showed consistency with established GWAS from the respective cohorts. Beyond loci, we showed high genetic correlation (0.78-1.0) for asthma, AFib, and RA between UKB and FinnGen although different phenotype curation approaches were used. In addition, a preliminary genome-wide survival analysis has been successfully conducted in one of the biobanks, anticipating the value of this approach in the network.

We use exemplar phenotypes to explore the impact of phenotype harmonization, imputation panel, and statistical methods towards the development of a more definitive, global resource for human genetics with mature analytic expertise.