The Observable Antibody Space (OAS) is a dataset of billions of antibody sequences and is widely used to train antibody language models. It was first published by (1).
1.
Kovaltsuk A, Leem J, Kelm S, Snowden J, Deane CM, Krawczyk K. Observed Antibody Space: A Resource for Data Mining Next-Generation Sequencing of Antibody Repertoires. The Journal of Immunology. 2018;201(8):2502–9. Available from: https://doi.org/10.4049/jimmunol.1800708