Shikeagi Kato, an endocrinologist formerly of the University of Tokyo who resigned on March 31 amidst an investigation into his work, has retracted another paper, this one in Nature.
Here’s the notice for “DNA demethylation for hormone-induced transcriptional derepression,” which was the subject of a correction last October:
In this Letter, we claimed that hormone-regulated transcriptional control involves DNA demethylation mediated by MBD4. Subsequently, we corrected some figure panels that appeared to have been erroneously prepared (Corrigendum, http://www.nature.com/nature/journal/v480/n7375/full/nature10604.html). However, we later found that other figure panels contained data duplications. After further review, we now conclude that the results presented in the original figures had been inappropriately manipulated, and given these more serious concerns we no longer have confidence in the original figures. We therefore wish to retract this Letter and sincerely apologise for any adverse consequences that may have resulted from the paper’s publication.
Those adverse consequences are likely to begin with headaches for the authors of 94 papers that cited the study, according to Thomson Scientific’s Web of Knowledge. (The citations via Thomson Scientific, not the headaches.)
The Corrigendum, by the way, ended with an all-too-common scientific version of “We’re sorry! But nothing to see here”:
Our results and conclusions are not affected by these errors, but we apologise for the careless mistakes made.
Apparently, the inappropriate manipulations — pardon us, “errors” — did affect the results and conclusions.
This is the fourth retraction by Kato we’ve reported on. The investigation into Kato’s work, as we noted in April:
…was prompted by an outside whistleblower’s allegations in January about 24 of Kato’s papers. The whistleblower claimed that the papers manipulated and reused data improperly, and created a YouTube video to spread the word, as ScienceInsider reported earlier this year.
Hat tip: Genevra Pittman
It is now clear that this was a case of scientific fraud, with the evidence building up that this lab engaged in fraud on a very large scale.
But there are wider questions.
Nature allowed the authors to make a Corrigendum that simply perpetuated the fraud. There appears to have been no attempt to oversee the process of that Corrigendum, for example by asking independent reviewers to examine the images to see whether the authors’ statements were credible. Such a process would have quickly and easily revealed that not only were the statements about the paper’s conclusions being unaffected not credible (image manipulation undermines confidence in all the data) but the Corrigendum did not even cover all the instances of image manipulation in this paper. Incredibly, after the Corrigendum it was still necessary for an external whistleblower (11jigen) to point out that there were in fact 20 separate instances of image manipulation in this single paper, a fact which seems to have been entirely missed by Nature’s editors. JBC showed the way with the Gopal Kundu case where the paper was retracted by the journal despite the authors’ insistence that in fact all was well. No one has sued JBC and the earth is still turning. How can Nature have been so unable, or unwilling, to do the right thing?
Here’s a simple question – how many times was this paper quoted between the Corrigendum and the Retraction? All of those instances of contamination of the literature are directly Nature’s responsibility. So the journal owes its readers an apology, as well as the authors.
When one the world’s top three journals cannot spot blatant fraud when it is sitting directly under its own nose, what hope is there for science? More specifically, why should the public now have any faith in anything published in Nature? These are questions that Nature now needs to answer, and soon.
Sincere respect to 11jigen for doggedly pursuing this, as well as to RW for bringing it to wider attention.
Here is a lead that should be followed. Of those 94 papers who cited this paper, how many “replicated” the initial findings? Is it that absurd to hypothesize that a subset of them also made up data? Think about it…
A good question.
And whatever the answer, this case confirms that the idea that large numbers of citations validate a paper’s conclusions is baseless – indeed is there any relation between the validity of these high-impact papers (judged by Correction / Retraction) and the number of citations? I suspect not, or that it might even be inverted (fraudulent papers having more citations).
To answer my own question above, this paper has been cited 17 times in 2012 (i.e. after the correction but before the retraction), thanks to Nature.
spot on. Empire was built upon such data…no one will refute because it was published in high profile journal – such data would be piled upon each other. Once this is unearthed only the original paper is retracted? What about the rest who supported the observation? It is not absurd to hypothesise Jerry – there is a tendency of supporting papers published in high profile journals. Same scenario for “spice” case..
And may God bless all who sail in her!
(FYI: ships and large boats are feminine in English. That is the traditionalist view).
Is this an (bad) example of high impact journals’ review system? How did it pass through the extensive review process? It would be nice if the journals reveal the reviewers of such papers (of course after retraction). I know many of you don’t like this idea. Sorry if I hurt anyone.
Reviewers do not necessarily look for fraud as they usually take all data in the manuscript at face value. The whistle-blower who uncovered Kato’s fraud has been very clever in this regard.
agreed Chirality. but empty lanes in the western blot attract some sort of doubts…but not for favourable reviewers.
This issue of……
“Our results and conclusions are not affected by these errors, but we apologise for the careless mistakes made”
….needs to be dealt with. There is no way authors who admit to “errors” in data presentation should EVER be allowed to make this statement. Aside from the questions over the original paper, the manuscript has been retracted and, therefore, it officially no longer exists. How can the results/conclusions still stand if the paper is gone? No paper = no data = no conclusions. It’s pathetic.
That statement was made in the *corrigendum*. The paper was retracted later.
The “our results still stand” bit is from the earlier correction, not the final retraction. But I agree that such comments should not be allowed.
Mi-Sun Kim, Korean 1st Author of the retracted paper in Nature, has been working in Baylor College of Medicine (Division of Diabetes, Endocrinology & Metabolism) as a postdoctoral associate.
http://bric.postech.ac.kr/hanbitsa/author_cvpop.php?id=12624&Board=hbs_treatise&ttype=0&idauthorid=5746
I wish somebody does statistics on which journal retracts most. That’s authors fault, but the journal should be more careful. Is it my imagination that Nature papers are too often retracted, even comparing to Science or Cell?
Several people have. Ferric Fang and Arturo Casadevall created a Retraction Index, which corrects for how many papers journals publish: http://www.retractionwatch.com/2011/08/11/is-it-time-for-a-retraction-index/
And here are the raw numbers: http://pmretract.heroku.com/journals
Thank you, seems like I have been imaging more. It seems like Cell and Science are retracting more… Anyway, something must be fatally wrong. I see no correlation between real impact of the paper and the “impact factor”. Often case, fancy journals are often buying “cheap solution” (clear question clear answer, —but sometimes it’s clear because it is fabricated to make sense). Well, I have to constantly fight against the urge to quit science.