Scientists from Harvard Medical School and the University of Oxford have developed an innovative AI tool named EVEscape, which has the potential to predict virus mutations before they occur. This groundbreaking development could revolutionize our ability to prepare for emerging viruses and create timely vaccines to protect populations.
EVEscape operates by combining an evolutionary model that anticipates potential virus alterations with extensive biological and structural information about the virus. This amalgamation enables the AI tool to forecast which virus versions are most likely to emerge in the future. The tool’s accuracy was put to the test by feeding it data only up until January 2020 and observing its predictions regarding mutations that indeed surfaced after that date.
EVEscape outperformed existing experimental methods in predicting the commonality of these mutations. Crucially, it doesn’t depend on the development of antibodies or laboratory-based testing, making its predictions both faster and more effective.
In addition to its predictive capabilities, EVEscape has the capacity to sift through the vast number of new SARS-CoV-2 variants that arise each week and identify those most likely to pose significant threats. The tool’s real-time application to track the evolution of SARS-CoV-2 is currently underway, offering a vital means to understand how the virus may change in the future.
“We want to know if we can anticipate the variation in viruses and forecast new variants — because if we can, that’s going to be extremely important for designing vaccines and therapies,” said senior author Debora Marks, professor of systems biology in the Blavatnik Institute at HMS.
“You can use these generative models to learn amazing things from evolutionary information — the data have hidden secrets that you can reveal,” Marks said.
The model’s usefulness in predicting new variants first came into play when the coronavirus started mutating.
“We underestimate the ability of things to mutate when they’re under pressure and have a large population in which to do so,” Marks said. “Viruses are flexible — it’s almost like they’ve evolved to evolve.”
Furthermore, EVEscape could be employed to evaluate vaccines and treatments against present and future viral variations. This adaptive approach allows for the development of medicines that can effectively combat the virus’s changing escape mechanisms, a crucial feature in a world where viruses continually evolve.
Noor Youssef, a research fellow involved in this endeavor, expressed the desire to “design vaccines and therapies that are future-proof,” emphasizing the significance of preparing for the ever-changing landscape of viral threats.
In essence, EVEscape represents a pivotal advancement in our ability to forecast and address virus mutations. By harnessing the power of artificial intelligence and evolutionary models, scientists are equipping themselves with a dynamic tool that can safeguard our populations and provide a proactive response to the evolving landscape of infectious diseases. This not only aids in the development of effective vaccines and treatments but also facilitates ongoing monitoring and adaptation to emerging viral challenges.