Send lawyers, Einstein and Maugham: Authors object to PLOS ONE retraction

Ming Zhou

Here’s a tale of a paper retracted because other articles published years later seemed to plagiarize it – and its unhappy authors, whose behavior the journal says hints at paper mill activity.

On January 16 of this year, Maria Zalm, a senior editor at PLOS ONE and team manager for publication ethics, asked the authors of a 2015 paper to respond to concerns about their work – which had been flagged on PubPeer the previous November – by February 6, according to an email seen by Retraction Watch. After some apparent back and forth, Zalm wrote to the authors on March 6 to say the journal had decided to retract the article.

One of the authors, Ming Zhou, responded on March 13 – the deadline Zalm had set for the authors to indicate whether they agreed with the retraction. Zhou, formerly of Cleveland Clinic and now chair of pathology at Tufts Medical Center in Boston, said he was “seriously concerned” about the journal’s decision. He continued:

1.      In your decision, you mentioned that “In this case we consider that retraction is warranted due to concerns about results presented in Figures 2, 3, and 4.”

Please consider the fact that this article was received by the journal on 12/18/2014, several years earlier than the other two articles which plagiarized several images from our article (one was received on 2/1/02017, the other one on 3/26/2019).  We had no relationship with the authors of the other 2 publications. We are therefore the victim, rather than the perpetrators,  of the other two fraudulent publications. If this is the main reason to retract our article, this argument is flawed and has no legal grounds.

2.      In your decision, you also mentioned that “Furthermore, during our editorial assessment, this article was identified as one in a series of articles for which concerns about potential manipulation of the publication process, specifically, the journal has concerns about similarities with other submissions in article content and apparent reuse of images.”

Please provide evidence of this accusation. I am happy to review it  for me to make a final decision.

Zhou closed by saying he is “the Chair of the Pathology Department at a major academic medical center in US. I rigorously defend the academic integrity and in no way condone scientific misconduct in any shape and form. However, we need to excise [sic] clear and objective minds when tackling the fraudulent academic misconducts.”

The same day, co-corresponding author Hongquan Zhang, of Peking University Health Science Center in China, responded to Zalm, also objecting and quoting Albert Einstein and Somerset Maugham:

We all disagreed with and felt very angry with your intended retraction of this paper. We did not do anything wrong in this study. The paper was done in my lab with collaboration with the US team when Dr. Huiying He, a pathologist, was working in the States. However, your team’s decision was based on just some pictures in our paper that are similar to or the same with some pictures used in the later publications that were in fact published 3 or 4 years late than us. Could you and your team believed that we can pass across several years to copy somebody else’s work based on Albert Eistein’s [sic] theory of superluminal speed? As a matter of fact, it is somebody else who plagiarizes our published data, but not we copied someone’s. I, here, strongly suggest you carefully revisit the papers and rethink the fact, and finally make a precise decision on this matter. An improper decision will hurt our team, your team and the reputation of PLoS One, except the ones who plagiarize other’s work. By the way, we have reported this matter to our institutional ethics committee for being improper treatment by PLoS One on our paper. We will also report this to our law office to protect our right on this matter. I am sorry that I may write to you a little harsh at this moment. However, does someone can count on people who are improperly treated speaking softly as the US [sic] novelist Maugham advised? In addition, we also sent emails to the corresponding authors of the two papers that contained the similar or the same data with us. We asked them to show their original pictures. In fact, we have those original data in hand. To this end, I suggest a thorough investigation before you make a decision. It will be all clear in the end. We have attached supporting files in this email, in which we showed our pictures in comparison with the other two. We also showed our original data with high resolution. It is already early morning in Beijing I could not translate our attached files into English. I am sorry I just attached these files in Chinese, and if you would like to read them please use Google to translate or otherwise we can translate them for you later if you still interest in reading them.

Zalm responded the next day:

Regarding the concerns with this article, as indicated in the previous decision letter this article was identified as one of a group of multiple articles that are connected by concerns about image reuse; this group of also includes multiple articles published between 2013-2016. In addition to the concerns regarding overlap in the published figures, the publication ethics team has raised additional concerns, including concerns with the extend of panel overlap in the underlying data provided for editorial review (50+ instances of (partial) panel overlaps were detected), concerns with the western blots provided as underlying data (the data provided for review were not uncropped and were insufficient to resolve the concerns with the western blot data), and concerns with the peer-review this article received (involving both concerns regarding undeclared COIs and the quality of the peer-review the article received). The concerns that connect the group of articles are such that PLOS suspects systematic manipulation of the peer-review process, which is sometimes also referred to as “paper mill activity”. As indicated in the PLOS Ethical Publishing Practice policies, in cases of suspected systematic manipulation of the publication process, PLOS may choose not to disclose detailed information about our assessment and specific concerns. 

In light of these concerns, some of which apply across a series of articles and cannot be resolved through discussions or revisions that only address your article, PLOS has determined that a retraction of the article is warranted. We understand that you are disappointed and possibly frustrated by this outcome. As indicated earlier, we have updated the author statement to indicate that you disagree with the retraction decision. If this information is incorrect, please let us know no later than 18 March 2024. 

On March 18, Zhang again raised his objections, to no avail. The paper — which was cited three dozen times, according to Clarivate’s Web of Science — was retracted April 2 with a notice that referred to it being tied to a group of nine articles “connected by concerns about image reuse.” Three of those papers had already been retracted by PLOS ONE.

The notice went on to state the concerns “affect Figs 2, 3, and 4,” noting similarities between various panels in those figures and two other articles, one published in 2018 in the Medical Science Monitor and one published in 2019 in Molecular Medicine Reports

The corresponding authors “stated they were unaware of the panel duplications,” according to the notice. “They provided data files to support the results presented in this article but the files did not resolve the concerns.”

Zhang and co-corresponding author Huiying He, also of Peking University Health Science Center, once again objected in an email on April 3, calling the decision “typical bias and hegemony.” Zhou also told Retraction Watch he is unsatisfied, saying the “journal so far has not addressed our chief concerns raised in the corresponding authors’ letters.”

David Knutson, a spokesperson for PLOS, told us the decision “is tied to series-level issues involving a group of articles and which cannot be resolved by discussions at the individual article level.” He continued:

As is explained in the retraction notice, this article was identified as one of a group of ten articles published between 2013 and 2019 that are connected by concerns about image reuse. We note that pone.0124338 was published in 2015, and that the duplications mentioned in this article’s retraction notice were published in 2018 and 2019. However, our investigation into the linked articles found concerns that cast serious doubt on the integrity and reliability of all papers in this group. Thus, in accordance with guidelines issued by the Committee on Publication Ethics (COPE) and supported by our policy on Manipulation of the Publication Process, we retracted all four PLOS articles involved. In line with this PLOS policy, we will not disclose specific details as to the editorial concerns.

Like Retraction Watch? You can make a tax-deductible contribution to support our work, subscribe to our free daily digest or paid weekly updatefollow us on Twitter, like us on Facebook, or add us to your RSS reader. If you find a retraction that’s not in The Retraction Watch Database, you can let us know here. For comments or feedback, email us at team@retractionwatch.com.

17 thoughts on “Send lawyers, Einstein and Maugham: Authors object to PLOS ONE retraction”

  1. Why not solely reference the alleged 2013 duplication in the retraction notice? This mess makes the editor look inept.

  2. the journal should refund the publication fee.
    we write
    we submit
    we publish
    we pay
    and they retract?
    how about the fee? they simply promote their journal.. they conduct the peer review and accept the paper.. now they retract because of the flaw.. in logic, the problem come from journal and editor.. proof the editorial did not peer reviewing properly

    1. Similar question here. I never understand where our publication fee goes. Journal “editors” almost do 0 editing. It’s the author who has to add/remove/reorganize contents based on reviewers’ comments, it’s also the author who has to ensure the citations/references are in the correct format, it’s still the author who has to make sure the article is written in fluent English. The reviewers are volunteers, most journals don’t have to spend money on printing hard copies, the cost for storing the electronic files of articles and images is very little compared to multi-media hosting websites. What is the cost of running a journal, actually?

      1. The “cost of running” these journals is flying the editor in chief (first class airfare and top-tier hotels) to quarterly board meetings at the publisher’s headquarters, and paying a “stipend” to the editor in chief.

      2. China flexing its muscles.

        I am against people pointing out problematic data in China for a start.
        It will help a thoroughly unpleasant regime. Do you really want a scientifically literate China?

        1. Valid point, however I still do it for a couple of reasons:

          1. Letting problematic papers pollute the literature reduces the credibility of all of the literature, regardless of the nationality of the authors.

          2. I think the PRC is embarrassed by the poor quality of much of the research they are paying for. I don’t mind embarrassing them some more.

      3. I thought about this a while back, and the numbers add up surprisingly quickly. Even though the bulk of the labor is free (reviewers, editorial board members etc.) there’s a fair amount of expense on the back end. Let’s say you’re a small journal that publishes 20 papers a month, and charges a flat $1500 OA fee. That’s $360k revenue pre-tax. That assumes no income from subscriptions (OA only). Processing those 240 papers from word docs into web format (clickable reference links etc.) and correctly-formatted PDFs would require 2 full-time staff, say $100k each (these folks are easily poach-able to the tech sector). Then healthcare, benefits, retirement plan, etc. Then you have office space rental, insurance, utilities and other expenses every business carries. Web hosting, site design & maintenance etc. would be another full time staff member. And you need VERY strong back-ups unless you want grief from the people who hand out DOIs and curate PubMed listings. No executive travel or other perks, just simple running costs and all the income is spoken for.

        Now of course, in reality there are economies of scale which make the above numbers laughable. That scale is what enables the big players (Elsevier, SpringerNature et al) to turn 40% profit margins on $billions in turnover every year. For example, it’s likely they’re using cheap overseas labor (Amazon Mechanical Turk) for the coding, and those folks aren’t getting retirement or healthcare benefits. And the web-hosting costs per paper drop when you own the server instead of renting storage.

        So, the cost of running a modern online journal is not zero, but it’s also not $1500 per paper. In-fact, based on that typical $1500 APC and those typical 40% profit margins, we could estimate it’s probably costing the big outfits about $1000 per paper. Anyone in publishing want to chime in and correct these assumptions with real per-paper costs?

        1. Processing those 240 papers from word docs into web format (clickable reference links etc.) and correctly-formatted PDFs would require 2 full-time staff, say $100k each (these folks are easily poach-able to the tech sector).

          Calling BS on this. Neither the workload, nor the salary makes sense.

          2 FTE workload = 320 man-hour per month = 16 man-hour per article. Given that even a small publisher has formatting templates easily available, isn’t the number over-exaggeratied by a factor of 5 at least?

          And with this workforce typically hired in South / Southeast Asia, can we be sure that the salary expenses are much in excess of $20k / year, even including employee-paid taxes?

  3. I observed the image reuse, but is there really a ground for retraction? or even a expression of concern? I doubt it. Even the presence of the image reuse, in the current context, wont affect the reliability of the reported results. Please refer to the manuscript’s text. The context is important. Perhaps PlosOne may perform more professional. As the ethics experts are not scientists they often make these mistakes and put the editorial under pressure for a retraction.

  4. Open access journals including PlosOne are often in rush to publish with minimum monitoring during peer-review, and PlosOne again in rush to retract with minimum consideration. Is it really that difficult to check a few figures prior to the publication. Please consider investing a tiny portion of your fund for improvement of the peer-review. And most importantly kindly check your reviewers’ and editors’ reliability.
    Last but not least, retractions must be considered on a case-by-case basis, not a group.
    As a researcher who cited this article, I, disagree with this retraction.

  5. PlosOne’s investigation must be transparent with clear result. Retracting all the articles is not a solution. They damage the reputations of those whom cited those articles. PlosOne is reliable for the damages caused to those who cited the retracted articles. We trusted PlosOne in the articles they publish and we were betrayed. Can Ms. Zalm kindly comment on this please.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.