Abstract
The complementarity-determining regions (CDRs) of antibodies are loop structures that are key to their interactions with antigens and are of high importance to the design of novel biologics. Existing approaches for characterizing the diversity of CDRs have limited coverage and cannot be readily incorporated into protein foundation models. Here we introduce ImmunoGlobulin LOOp Tokenizer, Igloo, a multimodal antibody loop tokenizer that encodes backbone dihedral angles and sequence. Igloo is trained using a contrastive learning objective to map loops with similar backbone dihedral angles closer together in latent space. Compared to state-of-the-art protein encoding approaches, Igloo can efficiently retrieve the closest matching loop structures from a structural antibody database, outperforming the existing methods on identifying similar H3 loops by 6.1%. Igloo assigns tokens to all loops, addressing the limited coverage issue of canonical clusters, while retaining the ability to recover canonical loop conformations. To demonstrate the versatility of Igloo tokens, we show that they can be incorporated into protein language models with IglooLM and IglooALM. On predicting binding affinity of heavy chain variants, IglooLM outperforms the base protein language model on 8 out of 10 antibody-antigen targets. Additionally, it is on par with existing state-of-the-art sequence-based and multimodal protein language models, performing comparably to models with 7 × more parameters. IglooALM samples antibody loops which are diverse in sequence and more consistent in structure than state-of-the-art antibody inverse folding models. We show that Igloo can rapidly and scalably prioritize functional antibody variants from large mutagenesis libraries, achieving a 1.9 × enrichment of experimentally validated HER2 binders in a zero-shot setting. Igloo demonstrates the benefit of introducing multimodal tokens for antibody loops for encoding their diverse landscape, improving protein foundation models, and for antibody CDR design.