Original title: Machine-learning-based Structural Analysis of Interactions between Antibodies and Antigens
Authors: Grace Zhang,Zhaoqian Su,Tom Zhang,Yinghao Wu
This article discusses the use of computational analysis to understand the interactions between antibodies and antigens, which can lead to the development of new treatments for diseases. While artificial intelligence has advanced the ability to predict protein-protein interactions, detecting the specific binding sites between antibodies and antigens remains a challenge. To address this, the researchers developed a deep learning model that can accurately characterize the patterns of interaction between antibodies and antigens. The model is able to distinguish antibody-antigen complexes from other types of protein-protein complexes with high accuracy. Furthermore, the model can identify antigens from common protein binding regions even if only the epitope information is available. These findings suggest that antigens have distinct features on their surface that antibodies can recognize. Additionally, the research indicates that one antigen can be targeted by multiple antibodies and that antibodies may bind to proteins that were previously unknown. Overall, these results support the precision of antibody-antigen interactions and highlight the potential for future advancements in predicting specific pairings. The authors have also stated that they have no competing interests.
Original article: https://www.biorxiv.org/content/10.1101/2023.12.06.570397v1