A Nature Cell Biology paper published only a few months ago by prominent researchers in Taiwan has sparked a heated discussion on PubPeer, which now includes a comment allegedly from an author saying they have requested its retraction.
Although a representative of the journal wouldn’t confirm to us that the authors had requested a retraction, the comment on PubPeer says the paper contains several figures that were “inappropriately manipulated” by the first author.
Here’s the full comment on PubPeer, tagged as coming from one of the authors of the paper:
As pointed out by peers, we found that several figures in the article were inappropriately duplicated by the first author, S-T CHA. Even if the main conclusion of the article is not affected by these flaws, the authors still feel that these misconducts have raised an issue on academic ethics, and believe it is necessary to retract the article. We deeply regret any inconvenience this event has caused for others.
The retraction letter was sent out to NCB chief editor on November 6th, 2016
A spokesperson for Nature Cell Biology told us about “G9a/RelB regulates self-renewal and function of colon-cancer-initiating cells by silencing Let-7b and activating the K-RAS/β-catenin pathway:”
Nature Cell Biology is aware of the discussions on PubPeer and is following an established process to investigate the issues. However, that process is continuing and we do not comment on individual ongoing investigations.
As reported by the Taipei Times:
The paper, which claimed to show how protease G9a could possibly regulate the proliferation of colorectal cancer cells and help scientists create new treatments, was trumpeted by the university in an online [National Taiwan University (NTU)] newsletter last month.
According to her Google Scholar page, Cha is based at NTU; we tried to reach her using variations of the NTU email address, but each bounced back.
Lin Ta-te, Chief Secretary at NTU, told us:
Cha Shih-Ting has resigned from NTU. The retraction of the paper by Cha et al. (2016) is currently under NTU investigation. The investigation includes Dr. Kuo.
Kuo is last author Kuo Min-liang, who holds an appointment at NTU, and is also a vice president at Kaohsiung Medical University in Taiwan. A number of his papers are being questioned in PubPeer which don’t include Cha as an author — including several whose co-author list includes the current president of NTU, Pan-Chyr Yang.
In 2008, as last author, Kuo issued a correction to a 2006 Cancer Cell paper on which Yang is the second author, citing image problems. Here’s the erratum for “The VEGF-C/Flt-4 Axis Promotes Invasion and Metastasis of Cancer Cells:”
In this paper, certain images in Figure 3C, Figure 5B, and Figure 6E were mistakenly used during preparation and assembly of the figures; the corrected figures are shown below. The expression of CNTN-1 and β-actin shown in the left middle panel of the original Figure 3C (H928/vector and H928/VEGF-C cells stably transfected with CNTN-1 siRNA2) were not the appropriate figures and should be replaced with the corrected Figure 3 here. Similarly, the expression levels of CNTN-1 and β-actin displayed in the original Figure 6E were incorrect and mistakenly used. The correct image is shown in the new Figure 6. We also regret that several inadvertent duplications took place involving the original Figure 5B and Figure 6D, as some of these images represented similar experiments analyzing primary tumors and lung metastases, respectively, using the same cell lines. This error occurred while submitting the revised manuscript, and the panels displaying immunoblots of tyrosine-phosphorylated Flt-4, expression of β-actin, and expression of VEGF-C were inappropriately replaced with images in Figure 6D. We have replaced all of the images in Figure 5B, which are shown correctly in new Figure 5. All of the corrected data are consistent with our previous conclusions; hence, the experimental results and conclusions of this article are not affected by these modifications. The authors apologize for the inconvenience caused by these errors.
We’ve contacted Kuo for more information.
Update 11/21/16 11:49 p.m. eastern: The paper has now been retracted.
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 daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.
Aside from the issues of alleged misconduct by the authors, at what point do we begin to hold journals accountable for publishing this stuff? This is not some podunk open access dump journal – it’s Nature effing Cell Biology! If we look past the messed-up situation where one of the most prestigious scientific publications on the planet is headed up by an EiC with precisely 5 career publications (one as first author), then we have to ask the bigger question… what should be the consequences for this screw up? Should heads roll at NCB for allowing this to get anywhere near the printed page?
Beyond individual job consequences for people who portray themselves as “gatekeepers” of scientific knowledge, the secondary question has to be about the disconnect between publishing companies sitting on piles of cash (35% profit margins on $billions in turnover), and yet they keep on screwing up like this. COPE just sits there mumbling about (non-enforcible) guidelines, the funding agencies appear powerless, and so the wheel turns.
In every other industry there are consequences when it is discovered you’ve been seling garbage (just look at VW’s diesel scandal for a recent example). Somehow academic publishing is immune to such scrutiny, and that has to change. We lack the equivalent of a Consumer Products Safety Commission for the academic publishing industry.
A determined con artist can create a plausible story. That’s why we must insist on open data – all data. One example: there must be a requirement to put unspliced gels in supplemental materials.
What the journals should be accountable for is quickly and transparently correcting the literature when they and the reviewers have been duped, not nonresponsiveness, delays and roadblocks.
COPE has guidelines that are usually simply ignored.
I agree Paul there needs to be oversight. You won’t get open data and accountability until the journals are forced to do it.
There are agencies that can do it – the funders.
NIH, NSF, Gates, ACS, etc. could and should set guidelines for publication that require open data and public access to raw results and datasets, except in very narrow circumstances (biodefense).
I agree with Paul Brookes (first comment above). We discussed this paper in our journal club a couple of months ago. While we did not spot any figure tinkering, we concluded that the story smelled fishy. Indeed several blots and data were for outside the expected when knowledgeable in the field. A more experienced editor would not have touched it with a stick.