AI Tool Aims to Predict Virus Variants before They Appear
Scientists have developed an artificial intelligence (AI) tool called EVEscape that aims to predict different virus versions, or variants, before they emerge. The tool estimates the likelihood that a variant will have the ability to escape the body’s immune defenses.
EVEscape works by analyzing collected data on the development and evolution of a virus over time. The system also takes into account detailed biological and structural information related to the virus. It is a form of “generative AI,” a technology designed to operate at human levels. The tool was created by researchers from the University of Oxford in Britain and Harvard Medical School in Boston.
Many viruses mutate in order to survive and thrive. This constant mutation creates new variants, some of which can spread more easily or become resistant to treatments or vaccines. The creators of EVEscape built upon their previous work with EVE, a tool used to study gene mutations causing human diseases like cancer and heart disease.
With the COVID-19 pandemic, the team saw an opportunity to repurpose EVE into a tool for predicting viral variants. They used the tool to predict the mutations that would develop from the SARS-CoV-2 virus causing COVID-19. The researchers reported that EVEscape accurately predicted the appearance and dominance of SARS-CoV-2 mutations. It also predicted which antibody-based treatments would become ineffective as the virus developed mutations to evade them.
Compared to traditional lab-based methods, EVEscape can make predictions more quickly and effectively. The tool does not need to wait for relevant antibodies to arise in the population and become available for testing. It is capable of examining tens of thousands of new virus variants each week and identifying the ones most likely to pose a problem.
Sarah Gurev, a student in the Electrical Engineering and Computer Science program at MIT and a leader of the research, highlighted the importance of quickly identifying the threat level of new variants to inform public health decisions.
The research team continuously uses EVEscape to analyze real-time data on the development of new variants of SARS-CoV-2. They share this information with the public and international agencies like the World Health Organization. The experiments also showed that the tool successfully predicted mutating behaviors for other common viruses, including HIV and the flu.
Debora Marks, a professor of systems biology at Harvard Medical School’s Blavatnik Institute and lead writer of the study, stated that they will continue to improve EVEscape’s ability to predict new variants for various viruses. The tool’s predictions can be crucial for designing vaccines and therapies.
The researchers’ findings were published in the journal Nature. This AI tool has the potential to revolutionize the way we track and predict the emergence of viral variants, enabling earlier public health interventions.