Data manipulation in a Cell Reports paper blew the importance of a kind of bacteria out of proportion.
Retracted this month — less than three months after it was published — the paper showed, according to a summary on the cover page:
B. subtilis is a symbiont that resides in the gut of C. elegans and generates nitric oxide that is essential for the host. Xiao et al. demonstrate that nitric oxide promotes defense against pathogenic bacteria by activating p38 MAPK, demonstrating the importance of commensal bacteria in host immunity.
But B. subtilis — a member of the Bacillaceae family — aren’t actually as plentiful as they appeared, explains the retraction notice for “Gut-Colonizing Bacteria Promote C. elegans Innate Immunity by Producing Nitric Oxide:”
Soon after online publication of our paper, we became aware of manipulation of data used to calculate bacterial abundances shown in Figure 1. Specifically, the finding that members of the Bacillaceae family of Bacteria were abundant in the gut of C. elegans is incorrect. Upon reanalysis, it was clear that the most abundant bacterial families are in fact Pseudomonoceae and Enterobacterioceace; Bacillaceae represented no more than 0.02% of bacteria in the gut of C. elegans. Given that the underlying data was not accurately represented in the figure, we believe that the most responsible course of action is to retract the paper. We deeply regret this circumstance and apologize to the community. All the authors have agreed to this retraction.
The article has not been cited, according to Thomson Reuters Web of Science. (That’s unsurprising, given its short run.)
A spokesperson for Cell Press, which publishes the journal, declined to comment on which authors were responsible for the manipulation.
The spokesperson told us that the problem was brought to light by concerned readers, and then:
Investigation of Figure 1 made it clear that a major conclusion of the paper was not supported, and the authors concluded and we agreed that retraction was the most appropriate course of action.
On PubPeer, commenters raised concerns about potential manipulation in a different image. The spokesperson told us:
We did look into Figure 4 as well, but the investigation that led to the decision to retract the paper was centered on Figure 1.
We have reached out to corresponding authors Cheng-Gang Zou and Ke-Qin Zhang (also the last author), who work at Yunnan University in China. We were unable to find contact information for co-first authors Yi Xiao and Fang Liu, also affiliated with Yunnan University.
Hat tip: Rolf Degen
Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our new daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.
Introduction of an international system for author identification with institutional email address (also after the author quit) and private phone number is highly needed. My experience is that when trying to contact especially Chinese authors, too many emails bounce back. It is also challenging to identify the individual Chinese authors, when over 100 million Chinese have the surname Li, and over 90 millions are named Wang.
The ORCID system (orcid.org) does just that, and journals could make use of that. ORCID can act on cases of fraudulent use, by removing those concerned or blacklist them. However, I would object to only allowing institutional e-mail addresses. The door must remain open eg to those who continue with scientific contributions after retirement.