Authors retract Science paper after investigation reveals manipulated images

Researchers at a prominent Japanese university have retracted a 2015 paper in Science, after an investigation uncovered image falsification and fabrication.

Last September, the University of Tokyo began an investigation of seven papers from the lab of cell biologist Yoshinori Watanabe after receiving anonymous allegations. In May 2017, the university determined that five papers contained falsified or fabricated images, and announced the results of its investigation on August 1. Two of the papers were published in Science, two in Nature and one in EMBO Reports.

On July 1 2017, EMBO Reports issued an erratum to the 2011 paper flagged in the investigation, correcting issues in several figures. Here’s the retraction notice for “The inner centromere–shugoshin network prevents chromosomal instability,” the 2015 paper in Science and the first of the papers to be retracted:

In the Report “The inner centromere–shugoshin network prevents chromosomal instability” by Y. Tanno et al., the representative images for comparison in Fig. 2C and figs. S13C and S15A were captured under inconsistent imaging conditions. In figs. S8B, S11F, and S12B, the images used for comparison were not adjusted using identical settings. In fig. S8A, the dot blots were mislabeled and not properly adjusted for contrast. The authors show corrected images and explanations as supplementary material. Given that they used the original images to calculate signal strength in the Report, the changes to the representative images do not affect the conclusions. However, because of the number of errors, the authors have decided to retract the Report.

The paper, which has been cited 15 times, according to Clarivate Analytics’ Web of Science, had already received a correction on Sept. 18, 2015, to replace two figures.

We asked the journal whether it plans to issue a retraction or correction to the other Science paper flagged in the report, “Two histone marks establish the inner centromere and chromosome bi-orientation.” A spokesperson told us:

The Science editorial staff and the authors have been in contact concerning lesser errors in a figure in the 2010 publication “Two histone marks establish the inner centromere and chromosome segregation,” and will be issuing a correction, which is currently under discussion.

That paper, published in October 2010, has been cited 194 times so far.

According to ScienceInsider, Watanabe has acknowledged deviating from best practices when dealing with images, but he has maintained that the issues flagged in the investigation do not alter the “main conclusions of any of the reports.”

Here’s the correction notice from EMBO Reports for the 2011 paper “Acetylation regulates monopolar attachment at multiple levels during meiosis I in fission yeast:”

In Figure 2A, due to over‐contrasting of the top two Western blot panels (AcPsm3), there is a loss of signal, in particular in the top right panel. Further, the right two panels were derived from the same blot, which was stripped and reprobed. The left two panels were derived from different blots of the same sample. This limits the quantitative information content of the published figure, but it does not change the conclusions derived from the experiment. In the corrected figure, the two left panels are now assembled from the same blot that was stripped and reprobed, and the contrast is properly adjusted in all panels. All these blots were processed in parallel in the course of the same experiment.

The paper has been cited 13 times.

Two Nature papers flagged in the report have not yet been retracted or corrected:

• “Heterochromatin links to centromeric protection by recruiting shugoshin” published in August 2008 and cited 101 times so far.
• “Condensin association with histone H2A shapes mitotic chromosomes” published in June 2011 and cited 103 times so far. According to the change history, on June 12, 2011, the “Supplementary Information PDF was replaced.”

We emailed Nature to ask whether the journal intended to retract or correct either paper and will update if we learn more.

In addition to the papers flagged in the investigation, two other papers have been questioned on PubPeer—a 2010 paper in Nature Cell Biology and a 2005 paper in Cell.

The University of Tokyo investigation examined allegations that six research groups at the university had fabricated and falsified data in 22 papers, though the university ultimately concluded that only Watanabe’s group had falsified and fabricated data.

In June 2017 – before the university issued its statement — Watanabe published his own statement to address what he called “errors” in the five papers in “the interests of transparency and a speedy resolution to concerns about these articles:”

As the head of the laboratory, I take ultimate responsibility for these errors, and extend my sincerest apologies to the scientific community for any concern or inconvenience these may have caused. Although we believe that none of the errors affect the main conclusions of any of the reports, we are in contact with the journals in which they were published to determine the most appropriate action we should take for each of the articles 

According to ScienceInsider, the university says it will decide on disciplinary action after reviewing additional papers from Watanabe’s group  And according to Nature News, all members of Watanabe’s lab have left and his 416-million-yen grant (US$3.7-million) from Japan’s science ministry was suspended last March. Watanabe’s lab website lists only one other member, Takeshi Sakuno, who was a co-author on three of the five papers flagged in the investigation.

Update, 820 UTC, Sept. 14, 2017: Watanabe provided his thoughts on the Science retraction:

Images meant for direct comparison should be captured and adjusted using identical settings. All errors in microscopic images happened because of unintentional mistakes in microscopy usage or in image adjustment. I deeply regret that these instances of non-identical acquisition settings or adjustment escaped my notice.

Watanabe also commented on the other four papers flagged in the investigation:

I admit that some of the image processing steps that we used were improper. We believe, though, that the improper processing of data in the other papers does not change the results, and these papers can be corrected using the original data. We are in contact with journals to correct these papers. For one paper (EMBO reports), a correction has already been published.

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.