Recombinant viruses formed by mash-ups of two variants of the SARS-CoV-2 coronavirus are now spreading from person to person, potentially increasing the risk of dangerous new variants arising.

New Scientistreported on the first detection of this kind of recombination last month, but at that point it was unknown whether the resulting hybrid was circulating in the wild. Two new analyses end any doubt. “Recombinants are circulating,” says Dave VanInsberghe at Emory University in Atlanta, Georgia.

Recombination is a potent source of evolutionary change in coronaviruses. The worry is that it could bring recent mutations together in new and more dangerous combinations, although there is no evidence yet of that happening.

In one analysis, VanInsberghe and his colleagues estimated that up to 1 in 20 of all SARS-CoV-2 variants circulating in the UK and US are now recombinants. The team analysed over half a million SARS-CoV-2 genome sequences from around the world and found more than 1000 possible recombinants. Most remain very rare, but two are circulating widely, one in the US, UK, Singapore, Japan and Canada, and the other in the US, UK, Canada and Denmark.

Read more: Two coronavirus variants have merged – here’s what you need to know

Neither of these two recombinants carry mutations that have been flagged up as being “of concern”, such as the ones seen in the variants first identified in the UK, Brazil and South Africa. “We have no reason to believe that the recombinants have altered transmissibility or virulence,” says VanInsberghe. Even so, he says, “these mark the first instances of widespread transmission of recombinants”.

Some of the other, rarer recombinants do carry those mutations of concern. “The real worry with recombination is that you recombine two lineages that have higher transmissibility or virulence, and that could be really dangerous,” says VanInsberghe.

Since the analysis was done, the most common recombinant has become even more numerous and widespread in the US, which could be a sign of greater transmissibility, he says. The research has yet to be peer-reviewed.

A separate analysis by the Walter Reed Army Institute of Research in Maryland looked at 100,000 genomes collected globally up to the end of October 2020, when fewer variants were circulating, and identified eight probable recombinants.

The research is also available online as a preprint. It is the institute’s policy not to comment on research that has yet to be peer-reviewed, but the paper says that the circulation of SARS-CoV-2 recombinants could have “major implications, especially if circulating recombinant results in escape from both natural and vaccine induced immunity”.

There is precedent for this, the researchers point out. Recombination between different strains of norovirus (not a coronavirus) have been shown to lead to rapid escape from naturally acquired immunity and lead to new pandemics of gastroenteritis.

“There is increasing evidence for recombination,” says Sergei Pond at Temple University in Philadelphia, Pennsylvania, who wasn’t involved in either study. He points out that, for now, recombination and regular mutation pose similar levels of threat, although that could change. “[Recombination] is not a major evolutionary driver at this point – recombinant strains are rare – but it will likely increase in prominence.”

References: bioRxiv, DOIs: 10.1101/2020.08.05.238386 and 10.1101/2021.03.07.434287