Retraction Watch

Tracking retractions as a window into the scientific process

Board member resigns from journal over handling of paper accused of plagiarism

with 12 comments

A biologist is crying foul at a journal’s decision to correct (and not retract) a paper he claims plagiarized his work — and one of his colleagues has resigned from the journal’s editorial board as a result.

The 2016 paper, published by Scientific Reports, is an application of a previously published algorithm designed to better identify regulatory sequences in DNA. The three authors, based at the Shenzhen campus of the Harbin Institute of Technology, used the technique to identify recombination spots in DNA. They called it SVM-gkm.

On April 2, Michael Beer of Johns Hopkins University in Baltimore notified the editors of Scientific Reports that he believed the paper had plagiarized his work. Despite Beer’s efforts, the journal ultimately decided to issue a correction notice, which cites “errors” and the authors’ failure to credit Beer’s work. That isn’t good enough for Beer — nor one of his colleagues at Johns Hopkins, who resigned from the journal’s editorial board saying “the recent affair with Mike Beer’s work being plagiarized did not impress me.

Beer claims that large swaths of the 2016 paper consisted of rewording his paper and restating the exact same equations that underlie his software (albeit with some errors introduced) —named gkm-SVM. And while his paper was referenced in the text as the original source of several (but not all) of the equations, Beer says wording in the abstract made stronger claims to novelty. Here’s an excerpt from the abstract of the 2016 paper:

To solve this problem, we add gaps into k-mer and introduce a new feature called gapped k-mer (GKM) for identification of recombination spots. By using this feature, we present a new predictor called SVM-GKM, which combines the gapped k-mers and Support Vector Machine (SVM) for recombination spot identification.

In effect, argues Beer, the authors had tried to pass off the work as their own.

On April 2 2016, Beer wrote to the journal:

This paper should be retracted based on the unauthorized reuse of figures, equation images, extensive text, and fraudulent claims of novelty.

In a follow-up email five days later, he also noted that the corresponding author of the paper, Bin Liu, was an editorial board member of Scientific Reports, and called for his removal from the board.

When we asked Bin Liu for comment, he wrote to us:

Obviously, our Scientific Reports paper “Recombination spot identification based on gapped k-mers” didn’t plagiarize any paper. I don’t need to say more about this, since the truth is already there: no journal will publish any plagiarized paper, especially for the famous ones, such as Scientific Reports. Our paper was objectively evaluated by SR, and this point is supported by the research community.

Scientific Reports is published by Nature Publishing Group as an open access journal that advertises itself as offering authors “Thorough peer review with a focus on scientific validity — rather than significance or impact.”

A controversial decision prompts fallout

On April 29, 2016, Scientific Reports’s managing editor Richard White notified Beer that the journal would issue a corrigendum—but not a retraction—and acknowledged the authors “overstated” their claims:

Although your work (including the PLOS Computational Biology) is cited in the Sci Rep paper, the authors’ claims about their contribution to the algorithm development are overstated and they should have made it clearer that they apply existing methodology to a new biological question…The inaccuracies and ambiguities in the paper do not warrant its retraction, but we will publish a corrigendum.

Beer pushed back, citing the wording in the abstract and the reproduced derivation. He asked White whether he considered this to be an instance of plagiarism. White responded on May 11 2016:

This is not a question of whether the paper qualifies as plagiarism, but a question of whether there is a level of plagiarism that warrants retraction. The policy link I sent previously explains that ‘minor plagiarism’ is relatively frequent; there is a wide spectrum of plagiarism, and not all cases of plagiarism warrant retraction.

Beer slammed the decision in an email to us:

If this is not a level of plagiarism that warrants labeling as plagiarism, I would like to see what is…I think this decision sends the message that it is acceptable to submit plagiarism to Nature.com’s Scientific Reports. If you are caught, you will be allowed to revise the plagiarized text once it is published.

Scientific Reports published its corrigendum on December 7, 2016 and amended the original paper with a revised abstract. The correction notice lists multiple changes, and begins:

This Article reports an application of methodology originally reported in Reference 33 to recombination spot identification. Reference 33 of this Article introduced a feature set called gapped k-mer for regulatory sequence prediction; this Article applied these gapped k-mer features to recombination spot identification, and a computational predictor was constructed for recombination spot identification.

In the original version of the Article, the Abstract included ambiguous sentences which failed to give due credit to the authors of Reference 33. The authors apologize for these errors.

One of Beer’s colleagues at Johns Hopkins who was a member of SRep’s editorial advisory panel, Aravinda Chakravarti, resigned following the case, citing the journal’s response as one factor in his decision. According to emails Beer shared with us, Chakravarti emailed Richard White on October 31, 2016:

I would like to resign from your Editorial Board and would like my name to be removed from the list. There are many reasons for my decision, including that I no longer have time to do the many things I accommodated before, but the recent affair with Mike Beer’s work being plagiarized did not impress me. I am sure you have your decision making process but I would rather have me spend time on my current journals…

Chakravarti told us:

I was deeply disappointed in SR’s judgement on this paper and am baffled why they would insist on publishing this manuscript.

He also said he felt the journal has deeper issues with its review standards that go beyond this case:

I agreed to join the board to help recruit junior editorial board members who would provide rigor in review. Over time, I noticed this was not the case and that I was involved in at least 2 reviews where the reviewers write to me after publication that they were surprised their major issues were not only not addressed but ignored. […]

I resigned because SR is not addressing these troublesome cases with intelligence. The question is not what the SR policies of plagiarism are appropriate but how they are enforced. This paper should have been rejected when the matter was pointed out by Mike Beer using other competent and independent reviewers. If it did pass review, which it did, then the ‘review system’ failed and I suggested that SR get a different new editor and ombudsman to look at the entire matter (manuscript, revision, correspondence between all reviewers and Mike Beer and me with SR) and render a judgement. At least SR would have shown a desire to address a serious allegation.

We reached out to Scientific Reports; a spokesperson for Scientific Reports sent us the following statement:

Scientific Reports carefully considers all concerns raised with the journal. In this case we looked carefully at the issues raised by the correspondent and involved independent peer reviewers in an assessment. As a result a corrigendum has been published which you can find here: https://www.nature.com/articles/srep35331.

Bin Liu is still listed on the editorial board of Scientific Reports; Aravinda Chakravarti is not.

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.

Comments
  • rfg October 10, 2017 at 9:30 am

    Professor Beer squarely hits the nail on the head:

    “I think this decision sends the message that it is acceptable to submit plagiarism to Nature.com’s Scientific Reports. If you are caught, you will be allowed to revise the plagiarized text once it is published.”

    This is the problem with allowing “corrections” for scientific misconduct.

    It’s not just NPG. Beer’s comment unfortunately can be applied in many other instances:

    “I think this decision sends the message that it is acceptable to submit to . If you are caught, you will be allowed to revise the once it is published.”

    This problem will continue at least until COPE revises its guidelines.

    A CORRECTION should only be used in the case of HONEST error, not to cover-up misconduct.

  • rfg October 10, 2017 at 10:40 am

    Correction (an honest error – I didn’t realize putting material between <'s gets it deleted):

    “I think this decision sends the message that it is acceptable to submit material with falsification, fabrication or plagiarism to publisher x's journal y. If you are caught, you will be allowed to revise the falsified, fabricated or plagiarized material once it is published.”

  • Benjamin Carlisle October 10, 2017 at 12:33 pm

    Unfortunately, this is not an isolated case.

    The Committee on Publication Ethics previously decided that this behaviour is acceptable, and that a correction is acceptable in cases of major plagiarism, even when errors are introduced:

    https://publicationethics.org/case/what-extent-plagiarism-demands-retraction-vs-correction

    COPE fails to recognise that authorship of a paper confers authority over its subject matter. This is why **retraction**, and not just correction in cases of plagiarism would best protect the scientific literature.

    This case (as well as the one I linked) illustrate this problem very clearly as the plagiarists have introduced errors to the methods. This failure to guard against the mis-attribution of ideas is a threat to the integrity of science.

    Unfortunately, this seems to have become a template for academic misconduct, as COPE and the journals involved have de facto given this behaviour their blessing.

  • Michael October 11, 2017 at 2:01 pm

    Scientific Reports publishes close to 70 papers a day. Anyone surprised that rigorous review might be hard to maintain?

    • ICC October 11, 2017 at 3:16 pm

      Certainly not me. This is the obvious outcome. And, it is driven by the fact that they charge £1,110 (UK and rest of world) / $1,675 (The Americas) at this time. Their profits are /not/ driven by quality peer-reviewers rejecting bad papers or by their ed. board retracting plagiarized work and setting the scientific record straight.

    • Marco October 12, 2017 at 1:54 am

      I don’t quite see this as a review problem. One might argue that a reviewer who had a good overview of the literature would immediately notice the paper was largely based on prior work of someone else. However, the literature is so vast these days, that this expectation would be naïve.

      The problem I see here is the substandard editorial response when the plagiarism is pointed out. Would the response have been any different if this was *not* a megajournal? Personally I doubt that, if the same people were involved in making the decision.

      • Miguel Roig October 12, 2017 at 12:17 pm

        I agree with Marco’s take on this issue, particularly the point that it is unreasonable to expect reviewers, even those who consider themselves to be ‘plagiarism freaks’ (those who actively look for plagiarism in much of what they read) to be able to easily recognize plagiarized work.

        (Plagiarism freak – A term I attribute to Mary Ellen Kerans, Author-editor based in Barcelona, Spain).

    • herr doktor bimler October 12, 2017 at 2:47 am

      Anyone surprised that rigorous review might be hard to maintain?
      Explaining, for instances, a paper on acupuncture.
      https://pubpeer.com/publications/4AC05B99D198296F99EBF7DAF926FE

      • Marco October 12, 2017 at 8:12 am

        I’d love to see a better comment on that Sci Rep paper than the Pubpeer entry, where the author seems to have stopped reading after the abstract and offers no substantial criticism of the actual experimental work.

        And if someone thinks it is all a problem of poor peer review, see this comment in Nature about the first reference in the Sci Rep paper, published in Nature Neuroscience:
        http://www.nature.com/news/2010/100602/full/465538a.html
        In particular the comment of Jana Sawynok is interesting, as it states that “caffeine blocks the adenosine receptor pinpointed in this study. Given the caffeine intake of many countries where acupuncture trials are carried out, this could be a serious confounding issue in trials”.

  • PJTV October 12, 2017 at 9:28 am

    There are 2 issues here
    1) It appears that this was not a verbatim copying of text by Wang et al, but the use of other people’s ideas of Beer et al. This is worse than simple plagiarism.
    2) How the journal, Scientific Reports, proceeded with this by only issuing a correction. The latter is subject of most comments.

    I wish somebody could clarify the first issue better. Has Wang et al made substantial contribution over or next to Beer et al?

    • Mike Beer October 12, 2017 at 12:36 pm

      There is an original contribution of the Wang et al paper, that is in applying the existing algorithm (gkm-SVM) to a new problem (recombination). The plagiarism consisted of claiming the development a new algorithm (SVM-GKM). The algorithm they used is exactly ours (gkm-SVM), and the software they used is exactly our software, released with the PLOS Comp Bio paper. To apply our algorithm to a new problem there is no need to copy our derivation and give it a new name, and it is not appropriate to claim “we present a new predictor called SVM-GKM.”
      –Mike Beer

  • DDam October 17, 2017 at 11:48 am

    Scientific Reports is a disservice to science. It uses NPG group to publish high-visibility ultra-low-quality papers. I’ve seen papers there that would flunk a lab report, let alone a peer-reviewed journal. Bold claims are getting promoted despite clear lack of evidence.

  • Post a comment

    Threaded commenting powered by interconnect/it code.