Creationism creeps in
A review entitled “Mitochondria, the missing link between body and soul: Proteomic prospective evidence” raised more than a few eyebrows when it appeared online in Proteomics on January 23. Jin Han from Inje University (South Korea) and Mohamad Warda from Cairo University set out to “merge the available scientific data related to mitochondrial proteomes and frame them into a reliable new agreement extending beyond the limited already accepted endosymbiotic hypothesis. . . .” In the midst of discussing the mitochondrial proteome, the relationships among mitochondria and other organelles, and the role of mitochondria in disease, the authors concluded, “More logically, the points that show proteomics overlapping between different forms of life are more likely to be interpreted as a reflection of a single common fingerprint initiated by a mighty creator than relying on a single cell that is, in a doubtful way, surprisingly originating all other kinds of life.” In short, creationism trumps the theory of evolution.
Michael Gray at Dalhousie University (Canada) says the provocative claim of having new proteomic data that refute the endosymbiont hypothesis “would really send my antennae waving.” He points out that Warda and Han presented no information or arguments to back up their claim. “Moreover, they misrepresent or are ignorant of the data that have accumulated over many decades in support of the hypothesis,” he adds. “I find it very difficult to know how it got through the [review] process. There is a tendency in some journals to really not pay much attention to review articles, and you can see where it leads in this instance.”
The scientific community launched a barrage of criticism after Paul Myers of the University of Minnesota Morris drew attention to the paper on his blog, Pharyngula (scienceblogs.com/pharyngula). “I think something went seriously awry in the review of this paper,” Myers says. “The peculiarities are so blatant I can’t imagine how it could have slipped past a reviewer.”
Thijs Ettema at Uppsala University (Sweden) agrees. “Given the unscientific—pro-creationist—nature of the paper, how could this paper have ever been published by such a high-standard journal?” he asks. “Was it even peer-reviewed?” The paper was peer-reviewed, including by someone who reviews frequently for Proteomics, the journal’s editor-in-chief, Mike Dunn, told the Chronicle of Higher Education (February 7, 2008, http://chronicle.com/daily/2008/02/1552n.htm).
“I view this as a breakdown of the scientific publishing system and not just the fault of a sloppy—and unpaid—reviewer,” says JPR’s editor-in-chief, William Hancock. “Except in this case the reviewer obviously did not even read the title of the paper, and so deserves the bigger share of the blame.”
The fact that the paper was even sent out for review mystifies Ralph Bradshaw, coeditor of Molecular & Cellular Proteomics. “If it were just two sentences in an otherwise fairly well written article, I would have said someone just skipped over it. It was a review, after all. But it’s pretty hard to argue that they didn’t read the title.”
Hancock says such a paper would normally be rejected in an editorial office. “All editors miss things and make mistakes, but . . . we, the scientific community, must be clear that this paper contained flawed and unscientific statements.”
How the paper slipped through the cracks is not known, because Dunn stopped giving interviews on the topic. Instead, the journal’s publisher, Wiley-VCH, sent out a press release on February 13 that said, “Clearly, human error has caused a misstep in the normally rigorous peer review process that is standard practice for Proteomics and should prevent such issues arising.”
But, Hancock says, “There are other issues, such as the value of a journal in terms of assuring product quality, and this is why the authors give control of their work to a journal. There is usually a significant improvement in quality by the review process—editor’s office, reviewers, and the production system—which should include well-trained copy editors and the scientific and public readership. In fact, the system still worked here but relied on the last step in the chain.”
Warda thinks the criticism is unwarranted. “As [in] any review article, our review article has to include our meaning about [a] certain known hypothesis that we agree with or disagree,” he says. “Therefore, we addressed those proteomic facts that agree with our meaning to let the reader share with us our assumption.”
As well as being criticized on scientific grounds, the authors were accused of appropriating parts of other investigators’ papers. “I had that reaction when I read it, because it is very uneven,” Bradshaw says. “My first response was, ‘Well, a lot of this paper is really quite reasonable.’ Then you begin to realize that there is something terribly wrong here—that you can have such polished sections intermixed with sections that were clearly not polished. And buried among those sections that were not so polished were some of these more astounding statements.”
On his website, John McDonald of the University of Delaware laid out evidence for the alleged plagiarism of at least six papers, including one of his own (http://udel.edu/~mcdonald/wardahan.pdf PDF Size: 140 KB). And Gray noticed similarities to his 1999 review in Science (1999, 283, 1476–1481). Warda says, “The paper has many borrowed sentences, which is why it was retracted from the journal.”
In an email to Myers that was posted on Pharyngula, Han wrote that he had asked the editorial office of Proteomics to retract the paper in “order to rectify an error.” The paper was retracted on February 12 “due to the overlap of passages with several previously published articles,” according to Wiley’s press release, which also acknowledged “the controversial viewpoints expressed by the authors.”
“Unfortunately, there was not a clear statement by the journal that the paper was retracted because the paper did not meet normal scientific standards deserving publication,” Hancock says. “As the editor of JPR, my feeling is the editorial staff of Proteomics should have acted promptly and decisively by retracting the paper as having been published in error for lack of scientific basis and plagiarism.”
Gray says it takes luck and some detective work to detect plagiarism. “However,” he adds, “there is no excuse for failing to pick up egregious and patently obvious deficiencies in logic, analysis, and argumentation, let alone practical issues such as mis-citation of the literature.”
“The paper by Warda and Han is, by several means, a scientific flaw, and should thus have never been published by any journal,” Ettema says. “Clearly, this is a call to reviewers to take [on] their responsibilities in keeping science as clean as a whistle!”
Myers isn’t convinced that editors will learn much from this cautionary tale. “Ideally, though, they’d learn that transparency is a valuable thing and that they don’t have a passive readership—what they publish is being read by critical eyes,” he says.
Bradshaw says there’s a lesson here for everyone. “Make sure you are very careful, when you are handling papers, that the reviews are thorough,” he says. “Any journal that thinks it is not at some point or other going to get stung with a flawed paper hasn’t been around long enough to understand the reality of it. . . . But to accept a paper like this, there has to have been a significant breakdown in the review process.”
“For the system to work—to catch all the plagiarism, the fabrication of data, and the unscientific claims—it has to be a shared responsibility among editors, reviewers, copy editors, and finally, the scientific public readership,” Hancock says, adding that the peer-review system is still the best available model, even though it is imperfect. And “in the end, science was upheld,” he concludes.
Advertisements
Return to Top
Copyright © American Chemical Society, 1155 Sixteenth Street N.W., Washington, DC 20036
