2021-09-30

Scientists use viral genome sequences to develop AI to identify high-risk viruses that potentially infect humans

By yqqlm yqqlm

Scientists use viral genome sequences to develop AI to identify high-risk viruses that potentially infect humans </ P >

<

p > identifying zoonotic diseases before they occur is a major challenge because only a small fraction of the estimated 1.67 million animal viruses can infect humans. In order to develop machine learning models using viral genome sequences, the researchers first compiled a data set of 861 viruses from 36 families. Then they established a machine learning model to assign the probability of human infection according to the pattern of virus genome. The authors then applied the best performing model to analyze the prediction model of zoonotic potential of other viral genomes sampled from a series of species</ p>

<

p > researchers found that the virus genome may have universal features independent of the virus classification relationship, and may make the virus adapt to infecting humans in advance. They were able to develop machine learning models that could use the viral genome to identify candidate zoonoses. These models have limitations because computer models are only a preliminary step in identifying zoonotic viruses that may infect humans. Viruses labeled by models will require confirmatory laboratory testing before significant additional research investment. In addition, although these models predict whether the virus can infect humans, the ability to infect is only part of the broader risk of zoonotic diseases. It is also affected by the toxicity of the virus in humans, the ability to spread between humans, and the ecological conditions at the time of human contact</ p>

Scientists use viral genome sequences to develop AI to identify high-risk viruses that potentially infect humans(1) </ P >

<

p > according to the study authors: “Our results show that the zoonotic potential of viruses can be inferred from their genome sequences to a large extent, which is surprising. By highlighting viruses with the greatest potential to become zoonotic, genome-based ranking can more effectively target further ecological and virological characteristics.”

Scientists use viral genome sequences to develop AI to identify high-risk viruses that potentially infect humans(2) </ P >

<

p > “these findings add a key part to the already surprising amount of information we extract from the virus’s gene sequence using artificial intelligence technology,” babayan added. “Genomic sequences are usually the first and often the only information we have about newly discovered viruses. The more information we can extract from them, the earlier we may determine the origin of the virus and the possible zoonotic risks it may bring. As more and more viruses are described, our machine learning model is identifying those that should be closely monitored and monitored It will become more effective in giving priority to the development of rare viruses for preventive vaccines. “