About this Research Topic
This Research Topic focuses on the recent advancements in computational methods for predicting antibody-antigen structure and interactions using machine learning, deep learning, and artificial intelligence approaches.
The binding mechanism between antibodies and antigens is crucial in structural biology and has important implications for drug discovery, immunological research, and vaccine design. Experimental techniques such as X-ray crystallography, NMR, and cryo-EM have led to the identification of 3D structures of virus-antibody complexes, making it possible to reliably identify antibody-antigen binding sites. Accurately evaluating the binding affinity between antibodies and a range of antigens is critical for developing effective antiviral treatments against global public health threats. However, predicting the structure conformation at antibody-antigen interaction sites remains a long-standing challenge for scientists due to the constant evolution of antigenic epitopes and changes in the 3D structures of paratope-epitope binding regions. This challenge is compounded by the time-consuming and expensive experimental solutions for antibody-antigen structure determination.
Recent developments in 3D structure prediction using deep learning, including AlphaFold series, RosettaFold, and RoseTTAFoldNA, have allowed researchers to better understand the biological functions of large biomolecular complexes. However, the predictability of antibody-antigen interactions without experimental techniques remains a challenge in the field. The algorithms currently available for predicting antibody-specific binding have limitations in identifying the amino acids that could bind to viral antigens without access to epitope information. This hinders their applicability in in vitro and in vivo applications. Therefore, it is essential to explore new computational methods to enhance the predictability of antibody-antigen interactions, which could enable the development of new therapeutics and diagnostics.
In order to tackle the research challenges surrounding antibody-antigen interactions as described above, we are proud to announce the organization of a special issue in collaboration with Frontiers in Bioinformatics. This special issue aims to present the latest and most exciting advancements in the field, focusing on the utilization of deep-learning-based structure modeling methods in conjunction with experimental data to comprehensively investigate the mechanism of antibody-antigen interactions. The articles featured in this issue will cover various aspects of antibody structure and antibody-antigen interactions prediction and determination, utilizing features such as sequence, tertiary and quaternary structures, binding affinity assessment, and analysis of experimental structural data.
We welcome the submission of computational methods for addressing the following problems:
• Computational analysis of antibody-antigen interactions
• Prediction of antibody structures
• Prediction of antibody binding sites
• Prediction of antibody-antigen interactions
• Analysis of antigenic epitope using experimental structural data (e.g., cryo-EM image data)
• Development of antibody-based therapeutics
Different article types including Original Research, Reviews, Perspective, and Methods will be considered.
The binding mechanism between antibodies and antigens is crucial in structural biology and has important implications for drug discovery, immunological research, and vaccine design. Experimental techniques such as X-ray crystallography, NMR, and cryo-EM have led to the identification of 3D structures of virus-antibody complexes, making it possible to reliably identify antibody-antigen binding sites. Accurately evaluating the binding affinity between antibodies and a range of antigens is critical for developing effective antiviral treatments against global public health threats. However, predicting the structure conformation at antibody-antigen interaction sites remains a long-standing challenge for scientists due to the constant evolution of antigenic epitopes and changes in the 3D structures of paratope-epitope binding regions. This challenge is compounded by the time-consuming and expensive experimental solutions for antibody-antigen structure determination.
Recent developments in 3D structure prediction using deep learning, including AlphaFold series, RosettaFold, and RoseTTAFoldNA, have allowed researchers to better understand the biological functions of large biomolecular complexes. However, the predictability of antibody-antigen interactions without experimental techniques remains a challenge in the field. The algorithms currently available for predicting antibody-specific binding have limitations in identifying the amino acids that could bind to viral antigens without access to epitope information. This hinders their applicability in in vitro and in vivo applications. Therefore, it is essential to explore new computational methods to enhance the predictability of antibody-antigen interactions, which could enable the development of new therapeutics and diagnostics.
In order to tackle the research challenges surrounding antibody-antigen interactions as described above, we are proud to announce the organization of a special issue in collaboration with Frontiers in Bioinformatics. This special issue aims to present the latest and most exciting advancements in the field, focusing on the utilization of deep-learning-based structure modeling methods in conjunction with experimental data to comprehensively investigate the mechanism of antibody-antigen interactions. The articles featured in this issue will cover various aspects of antibody structure and antibody-antigen interactions prediction and determination, utilizing features such as sequence, tertiary and quaternary structures, binding affinity assessment, and analysis of experimental structural data.
We welcome the submission of computational methods for addressing the following problems:
• Computational analysis of antibody-antigen interactions
• Prediction of antibody structures
• Prediction of antibody binding sites
• Prediction of antibody-antigen interactions
• Analysis of antigenic epitope using experimental structural data (e.g., cryo-EM image data)
• Development of antibody-based therapeutics
Different article types including Original Research, Reviews, Perspective, and Methods will be considered.
Keywords: Machine Learning, Deep learning, protein structure prediction, antibody-antigen interaction
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
