The authors of an influential but controversial 2020 paper on the activity of bat coronaviruses in China which proposed the animals as a “likely origin” for the virus that causes COVID-19 have retracted their work and republished a revised version of the analysis. They say their results and conclusions did not change.
The paper, “Origin and cross-species transmission of bat coronaviruses in China,” appeared Aug. 25, 2020, in Nature Communications. It has been cited 154 times, according to Clarivate’s Web of Science, and by at least two international policy documents.
The authors are affiliated with the Wuhan Institute of Virology and the New York City-based nonprofit organization EcoHealth Alliance, which has come under intense scrutiny by members of the U.S. Congress and others. The U.S. government in May suspended funding for EcoHealth amid concerns the COVID-19 pandemic virus may have developed from research on which the nonprofit and Wuhan lab collaborated – a so-called “lab leak.” EcoHealth has denied the pandemic virus could have emerged from its work.
Peter Daszak, a corresponding author of the paper and president of EcoHealth, told Retraction Watch the researchers had discovered dozens of the more than 1,000 viral sequences in the paper didn’t come from China, but from “a few miles over the border” in Laos. He explained:
Given that bats on both sides of the border likely co-mingle, and that our initial analyses were conducted by zoogeographic regions which encompass cross-border populations, we did not expect that the inclusion of these samples changed any conclusions substantially. However, given that the title and focus of the paper was for CoVs from bats in China, the authors decided to remove the 41 Laotian sequences, and repeat all of the paper’s analyses. During this process, we were also able to remove 29 sequences that were duplicated in the Genbank database. As expected, the results did not change substantially, and the conclusions remain the same.
Nature Communications published the revised analysis on December 19, along with a retraction notice for the original article. The abstracts of the original and revised papers are identical, except for the number of viral sequences.
The authors learned of the erroneously included sequences “months after publication,” Daszak said. He did not immediately respond to our questions about when his group informed the journal, or whether they considered publishing an expression of concern while they conducted the new analysis.
Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].