Antibodies are an increasingly attractive biological drug and currently at the forefront of the race against COVID-19, a pandemic that has caused over 1.1 million deaths as of October 20201–3. Therapeutic antibodies are currently generated experimentally, but structure-based rational design offers additional insight and benefits, most notably epitope specificity. So far, attempts at in silico rational design depend on experimental validation, suggesting that current computational methods do not account for all affinity-determining aspects of antibody-antigen interactions.My project involves studying antibody-antigen complexes, with the aim of better understanding these interactions. We hope this will guide development of better scoring methods for antibody models and be a valuable contribution towards improvement of computational design and understanding of antibody-antigen interactions.