Accelerating antibody discovery and design with artificial intelligence: recent advances and prospects

Ganggang Bai,Chuance Sun,Ziang Guo,Yangjing Wang,Xincheng Zeng,Yuhong Su,Qi Zhao,Buyong Ma
DOI: https://doi.org/10.1016/j.semcancer.2023.06.005
IF: 17.012
2023-06-01
Seminars in Cancer Biology
Abstract:Therapeutic antibodies are the largest class of biotherapeutics and have been successful in treating human diseases. However, the design and discovery of antibody drugs remains challenging and time-consuming. Recently, artificial intelligence technology has had an incredible impact on antibody design and discovery, resulting in significant advances in antibody discovery, optimization, and developability. This review summarizes major machine learning (ML) methods and their applications for computational predictors of antibody structure and antigen interface/interaction, as well as the evaluation of antibody developability. Additionally, this review addresses the current status of ML-based therapeutic antibodies under preclinical and clinical phases. While many challenges remain, ML may offer a new therapeutic option for the future direction of fully computational antibody design.
oncology
What problem does this paper attempt to address?