SARS- CoV-2 continuously mutates; each new variation often surprises the world. For example, the highly mutated Omicron variant emerged last November and required the health authority to develop a rapid response strategy despite initially having no answers to key questions: Protected How are people protected and how are people who have been infected before against this new variant? And are antibody therapies still effective against this new version of the virus?
Researchers led by Professor Sai Reddy from the Department of Biological Systems Science and Engineering at ETH Zurich in Basel have now developed a way to use artificial intelligence to answer such questions, potentially even in real time as soon as a new variant appears.
Discover countless potential variations
Because the virus mutates at random, no one can know exactly how SARS-CoV-2 will progress in the coming months and years and which variants will prevail in the future. In theory, there is virtually no limit to the ways in which a virus can mutate. And this is the case even when considering a small region of the virus: the mutant protein SARS-CoV-2, which is important for infection and detection by the immune system. In this region alone, there are tens of billions of mutations that could theoretically occur.
That’s why the new method takes a holistic approach: for each of these myriad potential virus variants, it predicts whether it is capable of infecting human cells and whether it is neutralized by antibodies produced by the immune system found in vaccination. those who have recovered. Most likely hidden among all these potential variations is the one that will dominate the next phase of the COVID-19 pandemic.
Synthetic evolution and machine learning
To establish their method, Reddy and his team used laboratory experiments to generate a large collection of mutant variants of the SARS-CoV-2 mutant protein. Scientists do not manufacture or work with live viruses, instead they only make part of the mutant protein, and so there is no risk of leakage in the laboratory.
The spike protein interacts with the ACE2 protein on human cells to cause infection, and antibodies from vaccination, infection, or antibody therapy work by blocking this mechanism. Many mutations in SARS-CoV-2 variants occur in this region, which allows the virus to evade the immune system and continue to spread.
Although the collection of mutant variants the researchers analyzed includes only a fraction of the several billion theoretically possible variants – this would not be possible to test in a laboratory setting. experiment – but it contains a million such variations. They carry different mutations or combinations of mutations.
By performing high-throughput experiments and sequencing DNA from millions of these variants, the researchers determined how successfully these variants interact with the ACE2 protein and with antibody therapies. now available. This tells us how well individual potential variants can infect human cells and how well they can get rid of antibodies.
