Researchers have developed an artificial intelligence tool that can forecast the emergence of new virus variants.
According to researchers from the University of Oxford and Harvard Medical School, the model could have predicted the evolution of covid-19 strains on mutation during the pandemic.
The model was christened EVEscape and should assist immunologists with designing new vaccine strategies based on how viruses adapt in order to survive a response to the human immune defense system.
Oxford University branded the tech as “predicting the future”.
The various variants developed through mutations caused different epidemic waves during the COVID-19 pandemic.
Such mutations can change how the virus behaves leading to ease of contagion or resistance to the immune system to clear it from the body.
However, it did not result into a significant increase in hospitalizations and fatalities, in late 2021 when the Omicron variant infected many people.
EVEscape or evolutionary model for variant effects incorporates a neural network for predicting evolution and specific structural information on a particular virus’s behavior.
The research team wrote in the paper of Nature how it works by stating that this mutation predicts the possibility that the virus will be able to overcome the blockade of antibodies and other immune responses.
University of Oxford and researchers Harvard medical school say it would be possible if this model was used to predict mutation of the Covid-19 during the pandemic.
The model called “EVEscape” is expected to provide useful information for the vaccine design and it also studies the mutation in HIV viruses due to the human immune system.
According to UK’s University of Oxford, it was “predicting the future”.
The wave during the covid19 pandemic was driven by various viral mutations.
Mutations in those genes can change how the virus behaves, perhaps causing it to spread more rapidly and allowing us to recognize and counteract it.
Towards the end of 2021, the Omicron variant did precisely the same as it infected millions, but without an overwhelming surge for admissions and fatalities reported.
The name “EVE” is an abbreviation for a complex evolutionary model that incorporates a deep neural network and comprehensive structural viral data.
The researchers described in Journal nature how it operates through forecast of probability that a viral mutant can evade antibody interference or other immune surveillance mechanisms.
The researchers tested the model using data available only at the beginning of the Covid-19 pandemic in February 2020. Impressively, it successfully anticipated which mutations of SARS-CoV-2 would emerge and become more common.
The team’s AI tool also foresaw which antibody-based treatments would lose their effectiveness as the pandemic evolved and the virus developed mutations to evade these therapies.
This technology holds promise for aiding in preventive measures and the development of vaccines that can target concerning variants before they gain prominence.
Pascal Notin, one of the study’s co-lead authors, mentioned that if this tool had been used at the outset of the pandemic, it would have accurately predicted the most prevalent Covid-19 mutations. He also highlighted the significant value of this work for pandemic surveillance and its potential to inform robust vaccine design in the face of emerging high-risk mutations.
Leave a Reply