Applications of Artificial Intelligence and Machine Learning toward Antibody Discovery and Development

Computational antibody design already employs established bioinformatic methods such as homology modeling, protein-protein docking, and protein interface prediction. Pharmaceutically focused computational methods support the assessment of antibody immunogenicity and biophysical properties. However, structure-based antibody design has been curbed by the lack of accurate antibody and antigen structures – until the emergence of machine learning (ML)-based methods capable of utilizing large data volumes. This chapter provides an overview of the applications of artificial intelligence and its subsets, ML and deep learning, in antibody discovery and development, covering the available databases and models for structure prediction, binding prediction, and developability, with special attention paid to the usage of language models for antibody design.