The collaborative efforts of researchers at Harvard and Oxford have yielded an artificial intelligence tool known as EVEscape. This tool is comprised of two key components: a sequence model that anticipates potential viral mutations and a repository of comprehensive biological and structural data about the virus in question.
(Photo : PHILIPPE LOPEZ/AFP via Getty Images)
TOPSHOT - Pupils wait to take Covid-19 antigenic tests available on the premises of the Fenelon Notre-Dame school complex in La Rochelle, south-western France, on January 13, 2022.
Predicting Virus Mutations Before They Happen
Researchers from Harvard Medical School and the University of Oxford have developed an AI tool called EVEscape. This tool has the remarkable capability to predict virus mutations before they actually happen, potentially revolutionizing our ability to prepare for, respond to, and develop vaccines against new viral variants.
Interesting Engineering reported that both institutions recently highlighted the groundbreaking potential of EVEscape. It combines predictive modeling of evolutionary sequences and extensive biological and structural data related to viruses to forecast potential mutations accurately.
Debora Marks, a senior author and professor of systems biology at the Blavatnik Institute at HMS, highlighted the research's primary objective: understanding the potential for predicting virus variations to revolutionize vaccine and therapy development.
She emphasized the ability of generative models to unlock hidden insights within evolutionary data, describing it as a tool for unveiling concealed secrets in the information. The AI model's role in anticipating new virus variants became evident during the coronavirus's mutation phase.
Marks underlined the remarkable adaptability of viruses under evolutionary pressures, emphasizing their capacity to undergo mutations, especially in highly populated environments. It's as if viruses have evolved to excel in the art of self-evolution.
Assessing Model's Precision
The modifications in virus behavior served as the catalyst for Marks and her research team to modify their novel AI tool for forecasting future coronavirus mutations. To assess the model's precision, Marks and her colleagues exclusively input data available up to January 2020 and witnessed its successful prediction of post-January mutations.
As stated on Harvard Gazette, EVEscape excelled at forecasting the prevalence of these mutations, surpassing the accuracy of conventional experimental techniques.
Moreover, its distinct advantage lay in its ability to make swifter and more efficient predictions than lab-based testing since it didn't rely on the development and availability of relevant antibodies within the population for testing.
This technology demonstrated the capacity to sift through the constant influx of tens of thousands of newly created SARS-CoV-2 variants each week, effectively identifying those with the potential to pose significant challenges.
Currently, the team is actively utilizing EVEscape to predict the future evolution of SARS-CoV-2, a critical effort in staying ahead of the virus.
Also Read: Can AI Improve Civility and Quality of Online Discussions? New Study Puts It to the Test
Furthermore, EVEscape's applications extend to the evaluation of existing and forthcoming viral mutations' impact on vaccinations and treatments. Axios reported that this capability empowers researchers to design treatments capable of withstanding the virus's evolving escape mechanisms.
As stated by Noor Youssef, a research fellow in the Marks lab, their objective is to pioneer the development of vaccines and therapies designed to withstand future viral transformations, ensuring a forward-looking approach to public health.
Related Article: AI System EWAD Can Provide Early Warnings for Dangerous Virus Variants, Scientists Says
