Oxford University develops AI tool to track virus variants

  • Published
Related topics
CovidImage source, Getty Images
Image caption,

The Omicron variant of the Covid-19 virus spread in late 2021

An artificial intelligence tool has been developed to predict new variants of viruses before they emerge.

Researchers at the University of Oxford and Harvard Medical School claim the model could have predicted mutations of the Covid-19 virus during the pandemic.

It is hoped the model, named EVEscape, will help in the design of vaccines by studying how viruses mutates in response to the human immune system.

The University of Oxford said the technology was "predicting the future".

During the Covid-19 pandemic waves were driven by different variants of the virus that had undergone multiple genetic changes.

Those mutations can alter the virus's behaviour, potentially making it spread faster or making it harder for our immune systems to recognise and fight off.

In late 2021, the Omicron variant did just that and infected millions, although it did not lead to a huge spike in hospitalisations and deaths.

EVEscape - short for Evolutionary Model of Variant Effect - combines a deep-learning model of how a virus evolves, along with detailed biological and structural information about it.

In the journal Nature, the research team described how it works by predicting the likelihood that a viral mutation will enable it to escape immune responses, for example, by preventing antibodies from binding.

'Tremendous value'

The model was tested with information only available at the beginning of the Covid-19 pandemic in February 2020 and successfully predicted which SARS-CoV-2 mutations would occur and which would become most prevalent.

The team said it also predicted which antibody-based therapies would lose their efficacy as the pandemic progressed and the virus developed mutations to escape these treatments.

It is hoped the technology could help in prevention measures and the design of vaccines that target variants of concern before they become prevalent.

Co-lead author for the study Pascal Notin, said it would have "accurately predicted" the most frequent mutations of Covid-19 if it had been used at the start of the pandemic.

"This work is of tremendous value, both for pandemic surveillance efforts, but also to inform vaccine design in a way that is robust to the emergence of certain at-risk mutations," he added.

Follow BBC South on Facebook, external, X, external, or Instagram, external. Send your story ideas to south.newsonline@bbc.co.uk.