Blood donor biobank pipeline to collect genome-based samples for research

The integration of genome data with electronic health records, driven by large biobank studies, has advanced human genetics by allowing systematic exploration of genotype-phenotype links. Regular donation enables large, longitudinal sample cohorts. Because blood donors are generally healthy, disease treatments or progression do not disturb interpretations in functional studies. We describe here a pipeline on how to collect blood donors' high quality plasma, serum, and living cell samples for multi-omics studies. Peripheral blood mononuclear cells (PBMC) were frozen and, after thawing, contained standard levels of immune cell subpopulations, responded to immune activation, and were of good quality starting material for single-nucleus multiome and cell imaging studies. We demonstrate that most genetic variants of interest to the major genomics study in Finland, FinnGen, could be found by random collection of samples during the standard blood donation without recall. Probing simple associations in the multi-omics data confirmed expected associations with e.g. age and sex, demonstrating good sample quality. As an example of interesting findings, we observed a significant association between frequent blood donation and lower levels of per- and polyfluoroalkyl substances (PFAS) (e.g., PFHxS β = -0.40 and p = 1.1 × 10^-18). The study demonstrates that regular blood donors are a suitable target population for high-quality, cost-effective sample collections.