The world is constantly learning how to live with COVID-19. But we are still under the horrors of a new species. New species with more infectious and ferocious dangers could turn the world back into a public health emergency. This includes a new city lockdown. and the resurgence of remote work and education. As a result, there is no shortage of jobs among researchers to design a COVID vaccine that can protect people from a wide variety of strains. Both now and in the future
Several institutions are racing to develop a universal COVID jab, and now MIT researchers think they might have found one that could take us across the finish line, if it did. do It ends up passing all the tests and getting into our arms. You will thank the AI.
A new study published in Frontiers in Immunology On March 9, scientists led by scientists at MIT reported the results of a successful trial vaccinating mice with a new vaccine designed to prevent them. all Strains of COVID—Even the Unseen Vaccines are being developed based on findings using AI algorithms, and a new trial demonstrates the vaccine’s ability to prevent deaths from COVID infection.
“This study offers proof of living systems. which is really a mouse “Vaccines we’ve invented using machine learning can protect against COVID-19,” said Brandon Carter, a doctoral student at MIT’s Computer Science and Artificial Intelligence Laboratory and lead author of the paper. The new paper said in a statement. The team characterizes its performance as the first experimental evidence that this AI-produced vaccine is effective against disease.
Today’s most popular COVID vaccines, such as those developed by Moderna and Pfizer, work by priming the immune system to identify the anti-Covid virus proteins. (a key component of how the virus initiates infection) and produces neutralizing antibodies Unfortunately, we’ve learned in recent years. The spike protein is also a structure that is more prone to mutations than other parts of the virus. It is these mutations that make strains such as Delta and Omicron particularly prevalent.
So researchers want to create a vaccine that works differently instead. Rather than activating vaccines to detect the COVID-19 virus early and expelling antibodies to prevent infection The team wanted to activate a portion of the immune system that produces and fires “killer” T cells that attack infected cells. This will not prevent infection. But it can be a very simple solution to stop serious illness or even death.
The team decided to incorporate AI techniques into the design process. It uses machine learning algorithms to identify which parts of the COVID virus should be used as part of the vaccine. The algorithm reduced the list of viral proteins from thousands to about 30. These are proteins that pop up on the surface of infected cells. And the immune system can determine what needs to be eliminated. The algorithm helps to determine which of these proteins are least likely to mutate with greater diversity.
After developing a vaccine The team squeezed lab rats and infected them with COVID-19 beta. The rats got sick, but they didn’t die. and they show a strong immune response due to vaccination. Some of the lung samples studied thereafter were 24 percent T cells.
That might sound like a low standard in meeting. And that is true to some extent. Most notably, the results proved that the vaccine works only with beta strains. But it’s still not a “universal” vaccine, but this baseline result means the vaccine meets the criteria for further testing in animals and eventually humans. It also opens the door to further investigation of T cell vaccines as a whole. This is especially true for immunocompromised individuals who are no longer able to produce neutralizing antibodies. and validate the use of new AI tools to help design therapies for new diseases.
