Nearly two and a half years into the COVID-19 pandemic, the emergence of new variants of interest of SARS-CoV-2 has prompted the development of a broad range of neutralizing antibodies. Variants such as Delta (B.1.617.2 lineage) and Omicron (BA.1 and BA.2) have been reported to exhibit immune evasion against some current therapeutic antibodies. Evolving SARS-CoV-2 requires rapid prediction of antibody binding to new variants and the development of broadly neutralizing antibodies.
A collaborative research effort between various medical science and engineering researchers from the University of Oxford (JianQing Zheng - NDORMS, Dr XiaoHang Fang - Engineering Science, Dr Zhu Liang - Nuffield Department of Medicine) and Nankai University (Professor HanTao Lou, previously also at the University of Oxford) took advantage of artificial intelligence and big data to develop and design an algorithm that can screen new coronavirus and its variant-specific antibodies.
The models developed in this study and later extended algorithms can address the discovery and optimization of antibodies for cancer, infectious diseases, and autoimmune diseases, thereby accelerating the development of therapeutic drugs
The developed neural network (XBCR-net) shows capability in predicting broadly reactive antibodies against newly discovered SARS-CoV-2 variants without prior knowledge of new variant-specific antibodies, enabling rapid generation of SARS-CoV-2 variants and other antibodies to emerging virus variants. In addition, the models developed in this study and later extended algorithms can address the discovery and optimization of antibodies for cancer, infectious diseases, and autoimmune diseases, thereby accelerating the development of therapeutic drugs.
In the context of widespread global vaccination against COVID-19, although the Omicron variant only causes asymptomatic to moderate clinical symptoms, it spreads very widely, and the case fatality rate remains high. At present, even if therapeutic antibody drugs can neutralize the new coronavirus and some variants, more than 70% of the antibody drugs are ineffective against the highly mutated variants of the coronavirus (such as Omicron). Therefore, we need to develop new therapeutic antibodies that can target all new coronavirus variants. Moreover, how to quickly respond to possible future variants of the new coronavirus and propose timely targeted treatments is also an urgent problem to be solved.
This work takes advantage of artificial intelligence and big data to develop and design an artificial intelligence algorithm that can screen new coronavirus and its variant-specific antibodies in high-throughput
Artificial intelligence-assisted biological research, especially immunotherapy, is booming and is now at the forefront of medical science and biomedical engineering research. Compared with traditional biological technologies, artificial intelligence has significant advantages in assisting protein structure prediction and target protein screening, which not only greatly reduces the time and cost of experiments but also improves the interpretability and quantitative analysis capabilities of biological problems. This work takes advantage of artificial intelligence and big data to develop and design an artificial intelligence algorithm that can screen new coronavirus and its variant-specific antibodies in high-throughput.
Different from the transformer-based deep learning algorithm, XBCR-net can support simultaneous input of multi-antigen antibody sequences for simultaneous screening, thereby predicting the binding probability of antibodies to multiple similar antigens. The research system is the first algorithm to predict the binding of antibodies to multiple antigens.
This work is of great significance for the prevention and control of the varieties that will appear in the future
This study is the first project of many led by Professor HanTao Lou and Professor XueTao Cao to screen broad-spectrum neutralizing antibodies using deep learning where the obtained broad-spectrum neutralizing antibodies have shown a very high affinity, proving the reliability of the research methodology. On this basis, the researchers found that the monoclonal antibody also has a high affinity for the SARS virus, suggesting that the neutralization ability of broad-spectrum antibodies against the SARS virus and new coronavirus and their variants can be improved through further antibody engineering optimization. This work is of great significance for the prevention and control of the varieties that will appear in the future. Researchers can also train XBCR-net with other diseases such as, HIV and influenza, for the discovery of neutralizing antibodies.
Image Credit: ShutterStock/Christoph Burgstedt
Showcase success for Science Together research
Public engagement with research