“Identifying the exact animal origin of the Kovid-19 virus can be a daunting task, predicting which innumerable animal viruses can emerge in humans, such as finding a particular grain of sand on a floor.“, Alert Australian researchers.

Still battling the endless Kovid-19 epidemic, many scientists are already looking at the best way to prevent such a health crisis again. But now, most tools to predict the next virus in humans are incomplete or unsuccessful, point these Australian researchers to the journal PLO Biology. For him, it is implicit in the increasing surveillance of all the places in which animals and humans interact.

Predictive and flawed predictive methods

Much work in progress is an attempt to predict the emergence of the next animal or human epidemic using statistical models. “We argue that the underlying data are often incomplete and will likely lead to biased and inaccurate predictions.“, Emphasized in the publication of three researchers from the University of Sydney (Australia).

The approaches used to assess zoonotic risk (the risk of passing the virus from animals to humans) fall into two categories. The “trait-based” approach identifies the characteristics of known zoonotic viruses, in terms of morphology, genetic evolution, or living environment. It then identifies other viruses that match this profile. The second approach, “network-based”, compiles pairs of known viruses and hosts, and points to other similar pairs.

The problem with these methods is that they are based on extrapolation, and favorable species for which great diversity has already been described. “Consequently, these methods may potentially have a tendency to predict livestock (and potential bats and rodents) as hosts.”, The authors’ conclusion.

Read in The research. Based on a forecasting model published in Nature communicationAt-risk animals include bats, pangolins, chimpanzees and African green monkeys, dromedaries, palm civets or common elephants in particular.

We do not even know 1% of all vertebrate viruses

Because despite the work of researchers around the world, the number of viruses known today is very small and probably does not even represent 1% of all vertebrate viruses. “Not only is our Virosphere sample extremely limited, but it is also strongly oriented towards viruses with socio-economic effects: which have an impact on human health, which are present in the species we eat or those we eat . As companions, and which cause significant and significant mortality events in domestic and wild animals“, Point to Australian researchers.

The Orthomexoviridae viral family, which includes all animal and human influenza viruses, is a good example. 97% of publications belonging to this family consider only two viral species: influenza A (including H1N1 for example) and influenza B (less common, infecting only humans only). Therefore, rather than predicting that viruses from families, genera, or species are more likely to emerge, “It is feared that many studies show that most studies have been done.“, Researchers regret. Thus,”Pangolins have been systematically overlooked as reservoirs of the virus, as it is now clear that these animals are infected with many viruses.“, They describe.

Monitor all places of contact between humans and animals

For them, the solution is to extract themselves from existing data in order to create new data, by monitoring in a targeted manner the places where the most interactions between humans and animals occur. “In addition to bats and people living around them, humans working in poultry farms, puppies, slaughterhouses and animal markets“Should be the subject of sampling and regular immunological surveillance, they argue.”This would provide a basic understanding of viral diversity in these potential hosts and provide an empirical, real-time estimate of the frequency of viruses spread between animals and humans, rather than an estimate based on biased and incomplete data.”