(ORDO NEWS) — The new model suggests that many more mammalian species than previously thought could become hosts for the new coronaviruses. While most coronaviruses that humans encounter usually cause only mild infections, the three most recent new strains – SARS-CoV-1, MERS-CoV, and SARS-CoV-2, which cause COVID-19 – are unusually virulent and have a relatively high mortality rate.
They also have a common origin story: they all evolved in other mammalian species. Interspecies transmission is one of the most common ways scientists learn about new viruses, but it’s also incredibly difficult to model and predict. Viral recombination complicates the process: when two different viruses infect the same cell and exchange genetic information to create completely new viruses. Variants of a virus can be created by nominal mutations in the genome of a single virus, resulting in small changes in the properties of the virus.
In a study published in the journal Nature Communications, biologist Maya Wardé and virologist Marcus Blagrove of the University of Liverpool developed a machine learning model to predict which mammals might host new recombinant viruses.
Recombinant viruses are viruses that require two different viruses to infect the same cell. The resulting virus can be very different from its parents. Coronaviruses are one of the few families of viruses that can recombine.
The model developed by the researchers predicted that compared to previous observations, this time the number of coronavirus-mammalian associations was 11 times greater. Moreover, the number of mammalian species that are likely hosts for coronavirus recombination has grown by more than 40 times. According to the authors of the scientific work, this indicates that the potential for generating coronavirus in mammals can be significantly underestimated.
Needless to say, the results obtained by the scientists are troubling. Today, science knows about the existence of 43 types of coronaviruses, but there may be many more. “We must learn to predict the origin of new coronaviruses in order to correctly focus resources and choose the right strategy of action, because when new coronaviruses spread to humans, it will be too late,” the authors of the study write.
Note that the model developed in the course of the study takes into account the genome sequences of known coronaviruses and mammals, as well as data on where mammals live geographically, what they eat and how closely related to other animals. In other words, the scientists managed to connect the existing biological data to the algorithm, that is, to teach the computer to recognize viruses and host species, which are most likely the source of recombination.
“We were able to predict which mammalian species could potentially infect other animals with coronaviruses,” explained Maia Varde in an interview with the BBC. “Either because they are very closely related to the carrier of the coronavirus, or because they share the same geographic space,” she said.
The big hurdle for the algorithm, however, is the lack of up-to-date knowledge of how certain viruses infect certain hosts. “We have very limited and biased data for a machine learning algorithm to learn from,” said Nicole Wheeler, a data scientist at the US Genome Pathogen Surveillance Center. “The mechanisms that determine whether a virus can infect a host and whether different viruses can mix are extremely complex and depend on factors that we do not have a lot of data on.”
Nonetheless, the new study represents a potentially transformative application of machine learning for disease surveillance and may help shed light on other hosts of the mammalian coronavirus that we should all be looking at. Recent studies have already confirmed several of the model’s predictions: the alpaca, domestic goat, and raccoon dog were found to be susceptible to SARS-CoV-2. So it’s wrong to blame bats or pangolins for the current pandemic. Most notably, the model identifies 102 potential hosts for the alarming recombination of SARS-CoV-2 with MERS-CoV, which is far more lethal than COVID-19.
According to virologists who were not involved in the study, the new work emphasizes the importance of timely detection of new coronaviruses, and also indicates that the number of potential mammalian hosts should not be underestimated. Whether we like it or not, animals, including humans, coexist with a large number of viruses. You can even say that humanity has been fighting viruses throughout history, and epidemics and pandemics have played a huge role in the development of our civilization. For this (and many other reasons), further research is very important, Vardo and Blagrov themselves have no doubt that the COVID-19 pandemic is not the last, which means that it is necessary to properly prepare for the next.
Contact us: [email protected]