Most new human infectious diseases come from animals, and scientists suggest they can be detected in time to save lives, with the right technology.
Recent reports have revealed that artificial intelligence may be able to detect the virus coming from an animal to a human before it becomes a pandemic.
Many scientists believe that the Corona virus, which continues to spread around the world, came from bats, in a process known as zoonosis.
Identifying high-risk viruses early can improve research and surveillance priorities.
Nardos Mullintz, Simon Babbian and Daniel Stryker of the University of Glasgow, who led the study, suggest that machine learning (a type of artificial intelligence) using viral genomes might predict the likelihood that an animal virus will infect humans, given the biologically relevant exposure.
The scientists explained that their study, published in the journal PLOS Biology, on Tuesday, September 28, may be a major breakthrough, as identifying zoonotic diseases before they appear is a major challenge because only a small minority of the 1.67 million animal viruses are able to infect humans.
To develop a machine-learning model using viral genome sequencing, the scientists first compiled a data set of 861 virus species from 36 families. Then they built machine learning models, which determined the likelihood of a human being infected based on patterns in the virus’ genomes. The team then applied the best-performing model to analyze patterns in the predicted zoonotic potential of additional virus genomes from a group of species.
“Our findings add an important part to the already amazing amount of information that we can extract from the genetic sequences of viruses using artificial intelligence techniques,” the scientists wrote in their paper.
The study indicated that the artificial intelligence developed by scientists, could have helped to identify “Covid-19” before it started killing in Wuhan, China, at the end of 2019.
“The ability to predict whether a virus can infect humans from just sequencing the genome, while still working reliably for completely new viruses that the model hasn’t seen, sets it apart from other methods,” co-scientist Mollentz told The Daily Beast.
Mullings and his team at the University of Glasgow helped research scientists at the University of Liverpool earlier this year looking at the potential of artificial intelligence in the field of human-animal viruses.
According to SciTechDaily, the scientists say: “Our results show that the potential of zoonotic viruses can be inferred to a surprisingly large extent from their genome sequences.”
“By highlighting viruses with the greatest potential for becoming zoonoses, genome-based classification allows for more environmental and viral characteristics to be targeted more effectively,” the team continued.
The team added: “The genetic sequence is usually the first, and often the only, information we have about newly discovered viruses, and the more information we can extract from it, the faster we can determine the origins of the virus and the animal risks it may pose.”
“As more viruses are characterized, our machine learning models will become more effective in identifying rare viruses that need to be closely monitored and prioritizing them for preventive vaccine development,” they wrote.
While the mechanistic models developed by the team predict whether viruses are able to infect humans, capacity to infect is only one part of the risk of spreading zoonotic diseases, which is also affected by virus virulence in humans, ability to transmit between humans, and environmental conditions in humans. time of exposure to a person.
Follow the health statement via Google News