AbGPT: De Novo Antibody Design via Generative Language Modeling

TL;DR

AbGPT is a novel generative language model tailored for de novo antibody design, capable of producing high-quality B-cell receptor sequences by leveraging advanced protein language modeling techniques.

Contribution

This work introduces AbGPT, a fine-tuned language model specifically designed for antibody sequence generation, advancing the application of protein language models in immunology.

Findings

Generated 15,000 high-quality BCR sequences.

Demonstrated understanding of antibody variability and conserved regions.

Showed potential for antibody library design and optimization.

Abstract

The adaptive immune response, largely mediated by B-cell receptors (BCRs), plays a crucial role for effective pathogen neutralization due to its diversity and antigen specificity. Designing BCRs de novo, or from scratch, has been challenging because of their complex structure and diverse binding requirements. Protein language models (PLMs) have shown remarkable performance in contextualizing and performing various downstream tasks without relying on structural information. However, these models often lack a comprehensive understanding of the entire protein space, which limits their application in antibody design. In this study, we introduce Antibody Generative Pretrained Transformer (AbGPT), a model fine-tuned from a foundational PLM to enable a more informed design of BCR sequences. Using a custom generation and filtering pipeline, AbGPT successfully generated a high-quality library of 15,000 BCR sequences, demonstrating a strong understanding of the intrinsic variability and conserved regions within the antibody repertoire.