Learning the Language of Antibody Hypervariability
A new transfer learning framework called AbMAP is proposed, which fine-tunes foundational models for antibody-sequence inputs by supervising on antibody structure and binding specificity examples, and provides compelling evidence for the hypothesis that antibody repertoires of individuals tend to converge towards comparable structural and functional coverage.
Abstract
Protein language models (PLMs) based on machine learning have demon-strated impressive success in predicting protein structure and function. However, general-purpose (“foundational”) PLMs have limited performance in predicting antibodies due to the latter’s hypervariable regions, which do not conform to the evolutionary conservation principles that such models rely on. In this study, we propose a new transfer learning framework called AbMAP, which fine-tunes foundational models for antibody-sequence inputs by supervising on antibody structure and binding specificity examples. Our feature representations accurately predict an antibody’s 3D structure, mutational effects on antigen binding, and paratope identification. AbMAP’s scalability paves the way for large-scale analyses of human antibody repertoires. AbMAP representations of immune repertoires reveal a remarkable overlap across individuals, overcoming the limitations of sequence analyses. Our findings provide compelling evidence for the hypothesis that antibody repertoires of individuals tend to converge towards comparable structural and functional coverage. We validate AbMAP for antibody optimization, applying it to optimize a set of antibodies that bind to a SARS-CoV-2 peptide and obtaining 82% hit-rate and upto 22-fold increase in binding affinity. We anticipate AbMAP will accelerate the efficient design and modeling of antibodies and expedite the discovery of antibody-based therapeutics. Availability:https://github.com/rs239/ablm