How to keep up with the Sars-CoV-2 coronavirus and its variants? Using artificial intelligence and machine learning to predict the risk of escaping the vaccine shield. By putting in place a team ready to design any updates for a recall. And testing the capacity of action of vaccine candidates on pseudoviruses constructed with the Spike protein mutations that characterize the most worrying variants of Sars-CoV-2. This is the ‘supply chain’ studied by Moderna. A platform that has already led to the development of a booster (mRna 1273.351) studied ad hoc to give an answer to the variants, starting from the South African one which had shown a “6-7 times reduction” in the neutralizing activity of antibodies. With this system, scientists from the American pharmaceutical company evaluated the protection achievable with the developed booster, and observed “excellent neutralization results across the spectrum, including the Indian variant”. This was explained by Guillaume Stewart-Jones of Moderna’s research platform, who with his colleague Melissa Moore addressed the issue of monitoring variants of concern, one of the topics at the center of the annual Science Day for analysts and investors. It starts from a maxi database that is fed with a cascade of data generated by multiple sources and allows continuous monitoring capable of promptly intercepting any warning signals to work on. To understand the amount of data with which one must measure, just think that “each ‘variant’ named generates a family of single ‘haplotypes’, each containing a unique combination of mutations”, explains Moore, showing a graph that shows, for example, the dense branching from the Indian variant. The question that arises at this point is: “What exact combinations of mutations should be incorporated into a new recall? The experts take the intercepted mutations and create pseudoviruses with which they can test the neutralizing activity of the antibodies generated by the vaccine. If this protection is reduced, a team intervenes to design any ‘updates’. This is how Moderna has developed an anti-variant booster. “Our data on the neutralization of the pseudovirus revealed that B. 1.351 “, the South African variant,” was a concern variant “to work on. The platform is in use and monitoring is continuous.” S We are building a highly efficient, data-driven and reliable system “to quickly identify variants of concern and be able to address a process of selecting ‘booster’ candidates of the anti-Covid vaccine for variants. Another challenge concerns the potential of the mRna science for the creation of “new classes of drugs for new targets”, as explained by the CEO of Moderna, Stéphane Bancel, assuring that “the best times” for this technology “are ahead of us, not behind us” . An example? Moderna scientists are targeting hematopoietic stem and progenitor cells in the bone marrow to modulate inflammation and immune disease.