MIT researchers have developed an AI tool called VaxSeer, designed to predict dominant flu strains and identify the most protective vaccine candidates months ahead of time. This tool uses deep learning models trained on decades of viral sequences and lab test results to simulate how the flu virus might evolve and how vaccines will respond.
VaxSeer has two core prediction engines: one estimates how likely each viral strain is to spread, and another estimates how effectively a vaccine will neutralize that strain. These engines produce a predicted coverage score, a forward-looking measure of how well a given vaccine is likely to perform against future viruses.
In a 10-year retrospective study, VaxSeer's recommendations outperformed the World Health Organization's (WHO) in nine out of 10 seasons for the A/H3N2 flu subtype and six out of 10 seasons for the A/H1N1 subtype. VaxSeer's predictions also showed strong correlation with real-world vaccine effectiveness estimates, as reported by the CDC and other health organizations.
The potential impact of VaxSeer is significant, as it could help health officials make better, faster decisions and stay one step ahead in the race between infection and immunity. While VaxSeer currently focuses on the flu virus's HA protein, future versions could incorporate other proteins and factors like immune history and manufacturing constraints.
By modeling how viruses evolve and how vaccines interact with them, AI tools like VaxSeer could revolutionize the way we approach vaccine development and distribution. As the researchers continue to refine and expand VaxSeer's capabilities, it has the potential to become a valuable tool in the fight against influenza and other infectious diseases.
