An artificial intelligence (AI) tool developed by a team from the University of Oxford and Harvard Medical School could help predict new viral variants according to findings in Nature in October 2023. The University of Oxford reports that EVEscape, the model, predicts the likelihood that a viral mutation will enable it to “escape immune responses”. EVEscape combines three sources of information to “score” individual mutations: 

  1. A deep generative model for fitness protection 
  2. Structural information about the spike protein to estimate antibody binding potential 
  3. Chemical distances in charge and hydrophobicity between mutated and wild type residues 

The research paper states that “extensive surveillance sequencing and experimentation” from the COVID-19 pandemic have presented a “unique opportunity” to assess EVEscape’s ability to “predict immune evasion” before escape mutations are observed.  

“The ability of EVEscape to identify the most immunogenic domains of viral proteins without knowledge of specific antibodies or their epitopes could provide crucial information for early development of subunit vaccines in an emerging pandemic.”  

Contributing author Associate Professor Yarin Gal believes that the “critical aspect” to the team’s approach is the methods “do not have to wait for relevant antibodies to arise in the population”. DPhil student and co-lead author of the study Pascal Notin commented that “had EVEscape been deployed at the start of the COVID-19 pandemic, it would have accurately predicted the most frequent mutations and the most concerning variants for SARS-CoV-2″. 

The study evaluated the model’s ability to make early predictions based on the limited information available at the start of the pandemic. It was able to successfully predict emerging and prevalent mutations as well as which antibody-based therapies would lose efficacy. EVEscape was also effective at predicting immune escape mutations for influenza, HIV, and “understudied viruses with pandemic potential” like Lassa and Nipah.  

Vaccine design 

The work represents a huge advance in disease control, and Notin suggests that it has “tremendous value” for pandemic surveillance and vaccine design. 

“The most exciting next step for this line of work is demonstrating how it can be used in practice to inform vaccine design.” 

Associate Professor Gal agreed that “anticipating viral variants that avoid immune detection with sufficient lead time” will be ‘key to developing optimal vaccines and therapeutics”.  

“Antibody escape mutations affect viral reinfection rates and the duration of vaccine efficacy.”   

How could EVEscape shape vaccine development for the future? For more on the technology behind vaccine design, subscribe to our weekly newsletters here.