A controversial plant biodiversity paper from 2014 was retracted yesterday, following more than 1 year of effort by a co-author turned whistleblower who sought to expose a range of data anomalies attributed to the paper’s senior author.
The paper purported to show that a method of distinguishing plant species by comparing snippets of DNA—known as DNA barcoding—is more cost-effective than traditional plant-survey methods. Ken Thompson, a postdoctoral fellow at Stanford University, worked on the paper as an undergraduate in 2014 at the University of Guelph (UG) with Steven Newmaster, a botany professor there. As noted by Science in June, Thompson later began to doubt the validity and provenance of the data, which Newmaster had supplied, and took the unusual step of asking UG and the Springer Nature journal Biodiversity and Conservation to investigate.
Thompson noted that the data had not been properly deposited to the GenBank database in 2014 as the paper stated, did not validate the paper’s conclusions, and seemed to bear a dubious similarity to separate data collected at the Canadian Centre for DNA Barcoding, also located at UG.
Paul Hebert, scientific director of the DNA barcoding center, in May voiced support for Thompson’s concerns and noted that the technique alone can’t reliably differentiate some closely related plant species.
The retraction notice in Biodiversity and Conservation corroborated several of Thompson’s concerns and said the editor “no longer has confidence in the validity of the data reported in this article.” Newmaster, it adds, “has not responded to any correspondence from the Editor or publisher about this retraction.”
Newmaster did not respond to a request for comment from Science. A UG spokesperson says the university “is not in a position to comment on a decision made by a journal.”
“I certainly am celebrating,” Thompson says, despite feeling odd about successfully seeking retraction of his first academic publication. “This is absolutely the correct outcome because it seems my co-author was unable to explain the anomalies in the data when given the opportunity to do so by the journal.”
Source: Science Mag