Covid breakthrough as new AI tool could help predict future viral outbreaks
The Covid-19 pandemic demonstrated how viruses are constantly evolving at a fast pace and have a way of evading our immune systems, which in turn, hinders any scientific efforts to control outbreaks.
New artificial intelligence developed by leading researchers at the University of Oxford and Harvard Medical School could help predict new variants of a virus.
The AI tool, named EVEscape, predicts if a viral mutation can escape immune responses in certain ways such as preventing antibodies from binding.
By AI understanding the biological and structural form of a virus, it is able to make predictions about what variant can occur when the virus evolves.
Co-lead author of the study, Pascal Notin, thinks that the AI tool could have helped fight the virus in the pandemic: “Our study shows 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.”
READ MORE: Doctor shares top four Covid symptoms seen in vaccinated patients
The scientists did an interesting test which involved only putting information into the model that was available to scientists at the start of the pandemic, back in February 2020.
It then asked EVEscape to predict what would happen with SARS-CoV-2 (the covid virus). The machine successfully predicted which mutations would occur during the pandemic and become the most prevalent in society.
The team believes it was also effective at predicting other patterns for influenza, HIV, and lesser-known viruses that they believe have the potential to cause a pandemic.
Ultimately, this information could help scientists to develop more effective COVID-19 vaccines and therapies.
Contributing author, Associate Professor Yarin Gal said: “We developed new AI methods that do not have to wait for relevant antibodies to arise in the population to predict which variants are the most concerning.”
The hidden wellbeing benefits of chocolate[LATEST]
NHS launches national pelvic health service to support women after childbirth[LATEST]
Dr Michael Mosley shares leftover pasta can slash your cancer risk[LATEST]
- Advert-free experience without interruptions.
- Rocket-fast speedy loading pages.
- Exclusive & Unlimited access to all our content.
The study, titled ‘Learning from pre-pandemic data to forecast viral escape’ has been published in the journal, Nature.
The EVE in EVEscape stands for ‘evolutionary model of variant effect’ and was first developed to predict the effects of genetic mutations on cancer or heart disease, but was then developed to predict viral variants.
The team is currently applying the AI tool to monitor Covid variants and sharing the data with the World Health Organisation to help prevent any future outbreaks of the disease.
Source: Read Full Article