MIT’s VaxSeer AI outperformed WHO flu vaccine picks in 9 of 10 seasons

MIT researchers have developed VaxSeer, an AI system that uses machine learning to predict which influenza strains should be included in seasonal vaccines months before flu season begins. The tool aims to reduce the guesswork in vaccine selection by analyzing decades of viral sequences and lab test results to forecast virus evolution and vaccine effectiveness.

What you should know: VaxSeer combines two prediction engines to forecast both viral dominance and vaccine effectiveness against future flu strains.

• The system estimates how likely each viral strain is to spread using a protein language model, then determines dominance by accounting for competition among different strains.
• A second engine estimates how effectively a vaccine will neutralize specific strains through simulated lab tests.
• Together, these produce a predicted coverage score that measures how well a vaccine will perform against future viruses, with scores closer to zero indicating better antigenic matches.

Why this matters: Current vaccine selection relies heavily on expert judgment made months in advance, often resulting in mismatched vaccines when flu strains evolve unexpectedly.

• When vaccine strains don’t match circulating viruses, protection drops significantly, leading to preventable illness and increased strain on healthcare systems.
• The challenge mirrors what scientists experienced during COVID-19, when new variants emerged just as vaccines were being deployed.

How it works: VaxSeer uses deep learning models trained on decades of viral data to simulate flu virus evolution and vaccine responses.

• Unlike traditional models that analyze single amino acid mutations independently, VaxSeer uses a large protein language model to understand the combinatorial effects of multiple mutations.
• The system plugs insights into mathematical frameworks based on ordinary differential equations to simulate viral spread over time.
• For antigenicity testing, it estimates vaccine performance in hemagglutination inhibition assays, which measure how effectively antibodies can prevent virus binding to human red blood cells.

In plain English: Think of VaxSeer as a sophisticated weather forecasting system, but for viruses. Just as meteorologists use historical weather data and computer models to predict storms, VaxSeer analyzes decades of flu virus information to predict which strains will dominate next season and how well vaccines will work against them.

Impressive track record: In a 10-year retrospective study, VaxSeer outperformed World Health Organization recommendations in most flu seasons tested.

• For A/H3N2 flu subtype, VaxSeer’s choices exceeded WHO performance in nine out of 10 seasons based on retrospective empirical coverage scores.
• For A/H1N1, the system outperformed or matched WHO recommendations in six out of 10 seasons.
• In one notable case for the 2016 flu season, VaxSeer identified a strain that the WHO didn’t select until the following year.
• The model’s predictions showed strong correlation with real-world vaccine effectiveness estimates from the CDC, Canada’s Sentinel Practitioner Surveillance Network, and Europe’s I-MOVE program.

Current limitations: VaxSeer focuses only on the flu virus’s HA (hemagglutinin) protein, the major influenza antigen.

• Future versions could incorporate other proteins like NA (neuraminidase) and factors such as immune history, manufacturing constraints, or dosage levels.
• Applying the system to other viruses requires large, high-quality datasets tracking both viral evolution and immune responses—data that aren’t always publicly available.

What they’re saying: Researchers emphasize the potential for AI to help health officials stay ahead of viral evolution.

• “By modeling how viruses evolve and how vaccines interact with them, AI tools like VaxSeer could help health officials make better, faster decisions — and stay one step ahead in the race between infection and immunity,” says Wenxian Shi, the study’s lead author and MIT PhD student.
• “Given the speed of viral evolution, current therapeutic development often lags behind. VaxSeer is our attempt to catch up,” explains Regina Barzilay, MIT’s School of Engineering Distinguished Professor for AI and Health.
• Jon Stokes, Assistant Professor at McMaster University, notes the broader implications: “Imagine being able to anticipate how antibiotic-resistant bacteria or drug-resistant cancers might evolve, both of which can adapt rapidly. This kind of predictive modeling opens up a powerful new way of thinking about how diseases change.”

Looking ahead: The research team is currently developing methods to predict viral evolution in low-data environments, building on relationships between viral families.

• The approach could extend beyond influenza to other rapidly evolving pathogens, potentially helping anticipate antibiotic-resistant bacteria or drug-resistant cancers.
• The work was supported by the U.S. Defense Threat Reduction Agency and MIT Jameel Clinic, with findings published in Nature Medicine.