and make them available, such as the effort by the Health and Retirement Study (https://hrs.isr.umich.edu/data-products/genetic-data/products#pgs). It also distinguishes our Repository from efforts to make publicly available PGI weights directly available for download28 (although we also do that, for weights constructed without 23andMe data). To deal with #3, for each phenotype and each dataset, we construct a PGI from GWAS summary statistics that excludes that dataset. We overcome #4 by using a uniform methodology across the phenotypes. In Methods, we detail how the Repository disseminates the PGIs, as well as the principal components of the genetic data in each dataset (which often should be used as controls for ancestry; see Supplementary Methods).
