The virus that causes COVID-19 could have started spreading in China as early as October 2019, two months before the first case was identified in the central city of Wuhan, a new study shows.

Researchers from Britain’s University of Kent used methods from conservation science to estimate that SARS-CoV-2 first appeared from early October to mid-November 2019, according to a paper published on Friday in the PLOS Pathogens journal.

The most likely date for the virus’ emergence was November 17, 2019, and it had probably already spread globally by January 2020, they estimated.

China’s first official COVID-19 case was in December 2019 and was linked to Wuhan’s Huanan seafood market.

However, some early cases had no known connection with Huanan, implying that SARS-CoV-2 was already circulating before it reached the market.

A joint study published by China and the World Health Organisation at the end of March acknowledged there could have been sporadic human infections before the Wuhan outbreak.

In a paper released this week as a pre-print, Jesse Bloom of the Fred Hutchinson Cancer Research Centre in Seattle recovered deleted sequencing data from the early cases of COVID-19 in China.

The data showed that samples taken from the Huanan market were “not representative” of SARS-CoV-2 as a whole, and were a variant of a progenitor sequence that was circulating earlier and spread to other parts of China.

Critics said the deletion of the data was further evidence that China was trying to cover up the origins of COVID-19.

“Why would scientists ask international databases to delete key data that informs us about how COVID-19 began in Wuhan?” said Alina Chan, a researcher with Harvard’s Broad Institute, writing on Twitter. “That’s the question you can answer for yourselves.”

Tweet from @Ayjchan

Serum samples still needed to be tested to make a stronger case about COVID-19’s origins, said Stuart Turville, associate professor at the Kirby Institute, an Australian medical research organisation who was responding to the University of Kent study.

“Unfortunately with the current pressure of the lab leak hypothesis and the sensitivities in doing this follow-up research in China, it may be some time till we see reports like that,” he said.

-AAP