A journal has retracted a 22-year-old-paper whose first author is the integrity officer for the Committee on Publication Ethics over concerns about image editing that “would not be acceptable by modern standards of figure presentation.”
The 2003 paper, “A recombinant H1 histone based system for efficient delivery of nucleic acids,” was published in Elsevier’s Journal of Biotechnology and has been cited 41 times, according to Clarivate’s Web of Science.
Sleuth Sholto David, who goes by the name “Mycosphaerella arachidis” on PubPeer, raised concerns about the image in December 2023, pointing out a “[d]ark rectangle” that appeared to be “superimposed onto the image.”
Coauthor Andrea Cristani, a professor at the Imperial College of London in England, responded on forum the same month, noting he was “not involved in in [sic] performing the experiments I have asked the person who produced the data to provide an explanation [sic].”
The first author of the article, Iratxe Puebla, is the facilitation and integrity officer for COPE, an organization publishers and journals often turn to for retraction standards. She is also the director for DataCite’s Make Data Count, an initiative “focused on open research data assessment metrics,” according to its website. The 2003 paper listed her affiliation at the time as Imperial College London.
According to the retraction notice, Puebla agreed with the decision and conceded “there appears to be evidence of splicing.” The “step would have been taken with the intention of providing a clearer presentation of the findings,” the notice reads.
Puebla did not respond to our multiple requests for comment sent to her work email, or to a message on LinkedIn. COPE said they are “are not able to comment on specific cases without having a thorough understanding of the facts” and did not respond directly to our request for additional ways to contact Puebla.
Christoph Sensen, the editor-in-chief of the Journal of Biotechnology, told us concerns about the paper were raised in January 2024 by a reader and investigated internally. He requested the retraction after authors on the paper failed to produce original data and declined to re-perform the experiment, he said. In a March For Better Science post, Leonid Schneider said he had contacted Sensen with concerns about the paper. Sensen did not say whether David or Schneider were the first to contact the journal.
Sensen attributed the discovery to advances in image technology. When the paper appeared, “it was printed in 8 bit. Now we have 16 bit and 32 bit imagery,” he said, which revealed the black box with two gel bands.
Sensen suspected the researchers “took two gels, one was darker than the other and they just cut out a piece from one gel and blotted it into the other one with Photoshop.”
“Even after 20 years, these things may just surface now which were hidden from sight before,” Sensen told us.
Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on X or Bluesky, like us on Facebook, follow us on LinkedIn, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].
” This step would have been taken with the intention of providing a clearer presentation of the findings, however they acknowledge that this would not be acceptable by modern standards of figure presentation . . . ”
So now 22 years-old articles are being retracted over issues with current “modern standards of figure presentation” that apparently did NOT exist at the time the article was published nor were considered a retractable offense? Wow. Just wow. If journal editors are going to impose “modern standards” (whether for figure presentation or otherwise) on decades old papers I’d imagine that you’d have to retract the vast majority of those decades old papers in the scientific corpus for violations of “modern standards” that simply did not exist back decades ago.
Perhaps an expression of concern outlining the problem with the figure would have been better than an outright retraction of a the journal article. But there you go and here we are.
William, that simply is not true, the authors were allowed their say, but they’re wrong. It never has been acceptable. Look at the image, it’s spliced together out of bits and bobs. It’s not an authentic image of a gel. They haven’t merely adjusted the brightness and contrast of that one area, because if they had, you’d still see the structure of the surrounding lanes. The two bands were transplanted in. Making stuff up has always been wrong in science.
The article clearly states that it “would not be acceptable by modern standards of figure presentation”, not that it “never has been acceptable”. This is pure revisionism leveraging on a minor flaw of the past generation with the intention to present the data better, not to present fake data.
Fraudulent images is not “revisionism of a minor flaw”.
A literal faked image, is called “a minor flaw”.
Faked image. I really hope you’re a bot or don’t work in science with that level of cognitive dissonance
More than sure you could pick any 25-30 years-old journal article out of any given journal and find something that would violate “modern standards” as they exist in 2025 that was perfectly acceptable practice decades ago. But does that mean they should be retracted? Nope, not in my opinion. Authors shouldn’t be punished for violating standards that didn’t exist at the time the paper was written and published.
I am not quite sure when superimposing only part of the lanes was the standard, I would be surprised that it was still acceptable in 2003. Even the first author seems to concede that this splicing is kind of fishy.
Kind of fishy when considered using “modern standards”. But would it be “fishy” considering the standards that existed in. 2003. That’s the question. In 2003 would the then existing publication standards make this kind of image manipulation a retractable offense.
The article here is erroneous in it’s presentation though: this is not about variable standards of what was acceptable but about an image the manipulation of which has now become clear due to differing computing standards. I’m pretty confident that at no point has it been acceptable to fake an image without acknowledging it, which is the issue reported and acted on.
You can clearly see they copy pasted those 2 bands in there. It’s 2 gels pasted together. Has always been fraud, and always will be. Don’t try to defend the non-defendable!
“. . . however they acknowledge that this would not be acceptable by modern standards of figure presentation . . .”
Then why did they use the the phrase “modern standards” implying that 22 YEARS ago this might have not been a problem for the authors or editors and that only by judging the figure using current “modern standards” that exist today would the manipulation would be a problem to the extent it would require a retraction.
So fraud that you could not detect X years ago is acceptable? Let’s cut the crap here. The picture is undoubtedly manipulated. There is no way you would not have seen it back then. They already had a figure file showing it (no matter how it was printed).
> Sensen attributed the discovery to advances in image technology. When the paper appeared, “it was printed in 8 bit. Now we have 16 bit and 32 bit imagery,” he said, which revealed the black box with two gel bands.
I do not follow. Clearly the original submission, be it hard-copy or digital, had 16- or 32-bit images, even if this was degraded to 256 grey levels when it was printed (then remastered for the digital edition). The editors and reviewers always had access to the same high-resolution images as Sholto David did.
I wonder if this explains COPE’s lack of enthusiasm for doing anything when there are integrity issues.
You do wonder if COPE is a publishers’ organisation. It does provide guidelines on what journals should do when a reader reports problematic data, but often the journals disregard the COPE guidelines, which state that the journal should thank the reader and come up with a plan, but ask additional questions, without reading, or looking at how the data might be problematic, want the reader to write some essay about the data, when the data are simply problematic, too similar, clash with other data, illogical…etc.
If course you can’t blame COPE for how some journals act.
I am getting a lot of ‘science guardians’ vibes here in some of the comments. Trying to defend bands from gel 2 that are pasted into gel 1 makes no sense. The reasoning that you shouldn’t use ‘modern’ techniques to showcase unethical or fraudulent practices in old papers is also a typical wrong reasoning often used by the science guardians.