Retraction Watch

Tracking retractions as a window into the scientific process

Author appeals retraction after co-authors dispute Nature Comm paper

with 21 comments

nature comm

Two weeks after Nature Communications published a paper on asymmetric cell division in July, it posted a retraction notice saying the paper was submitted “without the knowledge or consent” of all but the corresponding author.

The following day the journal “amended” the retraction note to include the initials of the corresponding author, Aicha Metchat, then based at European Molecular Biology Laboratory in Heidelberg, Germany.

The final notice for “An actin-dependent spindle position checkpoint ensures the asymmetric division in mouse oocytes” reads:

This paper has been retracted as it was submitted by the corresponding author A.M. without the knowledge or consent of the other researchers listed on the paper or their institution, utilising incorrect contact details. With the exception of the corresponding author, all other researchers have requested that the paper be retracted.

In response to an email, Metchat said the last author initiated the retraction:

Under the request of the last author, Jan Ellenberg, it has been taken the decision by the editor to retract my manuscript.

Ellenberg, at EMBL, corroborated this fact in an emailed statement:

In this case, as you saw in the retraction notice, a former postdoc from my group submitted a paper on her own account without consulting the researchers she listed as co-authors and without obtaining their consent or that of their institutions. Since she provided falsified contact information for all of them, the other researchers including myself, only became aware of it after online publication.

This went against EMBL’s above-mentioned rules and we therefore contacted the journal immediately requesting to withdraw the paper.

Metchat, however, said she has “started a procedure” with the Committee on Publication Ethics (COPE) to appeal the retraction, since she believes the paper doesn’t meet COPE’s guidelines for retraction:

Importantly, the nature of the retraction of my manuscript is going against the rules established by the Committee on Publication Ethics (COPE)…In the COPE retraction guidelines, it is specified that:

« journal editors should consider retracting a publication if:
• they have clear evidence that the findings are unreliable, either as a result of misconduct (e.g. data fabrication) or honest error (e.g. miscalculation or experimental error)
• the findings have previously been published elsewhere without proper crossreferencing, permission or justification (i.e. cases of redundant publication)
• it constitutes plagiarism
• it reports unethical research »…My case does not correspond to any of those criteria listed by the Committee on Publication Ethics. Therefore, I decided to do not request the retraction of my manuscript (Nat. Commun. 6:7784 doi: 10.1038/ncomms8784 (2015).

We also asked Metchat, who declined to disclose her current affiliation, how the authorship issue came about. She replied:

My case, it is an unfortunate mutual miscommunication:

To make long story short, I have been employed in the group of Jan Ellenberg from December 2008 to 2013. When, I finished my postdoc in December 2013, my work was ready to be published. Unfortunately, after working on the manuscript during three months with the group from Feb 14 to May 14, the principal investigator decided to interrupt the process of the publishing my work in June/July 2014 giving the priority to other projects in his group. Therefore, I decided to take over the preparation of my manuscript to apply for a fellowship in UK.

Retracting the paper — which has only been cited once, according to Thomson Scientific’s Web of Knowledge — hurts her career, Metchat added:

The decision to retract my manuscript is unfortunately only political decision. This will have disastrous consequences on my career progression.

In its online policy statement for authors, Nature says the following about author agreements:

Before submission, the corresponding author ensures that all authors are included in the author list, its order has been agreed by all authors, and that all authors are aware that the paper was submitted.

This policy was corroborated by Niki Scaplehorn, Chief Life Sciences Editor at Nature Communications:

The rationale for the retraction are set out in the retraction notice (http://www.nature.com/ncomms/2015/150729/ncomms9124/full/ncomms9124.html).  As we communicated to the author, since the publication of the retraction, we have indeed become members of COPE, which provides guidance on publication ethics. The COPE guidelines do not, however, replace our editorial policies, which remain in effect as they were when the decision to retract the paper was made and are clearly detailed on our website.  Submitting a manuscript without the knowledge or consent of the other researchers named is a clear violation of these policies (please see http://www.nature.com/authors/policies/authorship.html).  The licence to publish form that corresponding authors sign upon acceptance stipulates that the paper is submitted with knowledge and approval of all co-authors – please see points 4a and 4c http://www.nature.com/licenceforms/npg/mpl-ltp-cc-by.pdf.

We have no further comment at this time.

We also contacted Virginia Barbour, chair of COPE, who told us they recommend asking institutions to handle authorship disputes:

I won’t comment on this case specifically but our advice for anyone who has an issue with a journal is first to raise it with the journal and then if needed with the publisher. If it is due to disputed authorship we usually recommend that journal pass the dispute to by the authors’ institution(s) to handle and abide by that.

Metchat has said that she is planning to “create a blog in which I will explain my position about this question.” She also said that she has learned from COPE that they plan to “consult with members of the COPE Council” about her case.

Hat tip: Lee Thong Tan

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, and sign up on our homepage for an email every time there’s a new post. Click here to review our Comments Policy.

Comments
  • Some queries September 22, 2015 at 3:08 pm

    I have 4 queries:

    1) “the principal investigator decided to interrupt the process of the publishing my work” and “The decision to retract my manuscript is unfortunately only political decision”. Metchat, are these comments linked and can you provide a stronger rationale while you believe you were victimized, or why you feel this was a political decision.

    2) What does Matchat have to say about the NC author agreement?

    3) Can Matchat please elaborate on the “procedure” started with COPE. I am not aware that individual scientists can engage COPE for their own personal agendas in complaints against COPE member publishers, unless they themselves are COPE members.

    4) “since the publication of the retraction, we have indeed become members of COPE” This is somewhat strange for me. Why did NC only become a COPE member after the paper was retracted, and not before it? Were its ethical values all that different before the retraction, or did their publishing ethical standing fortify after becoming a COPE member? Understanding the reason and timing of this decision is essential for scientists like me who cannot comprehend why a journal’s ethical stance should be stronger simply because it is a COPE member.

    • Richard Telford September 22, 2015 at 4:39 pm

      On point 3, itis possible, though uncommon, for authors (or other parties) to take a complain to COPE if they believe that the publisher is in breach of the COPE code of conduct, but first the authors need to exhaust the complaints procedure at the publisher.

  • A Nawn September 22, 2015 at 3:44 pm

    Most people in the sciences always have “the one that got away”: something paper-able that just didn’t make the cut and never makes it into paper form (which is a waste of its own, but we’ll discuss that later). Unfortunately, forcing the submission and lying all the way through is not likely to demonstrate your integrity to a institution, or a funding agency, or a journal, all of which need to be convinced that you are faculty material.

    If she made the cut without the paper, she could probably discuss the unpublished work in her job talk. But if a Nat Comm. is pivotal to getting those job talks…then either the rest of her record is more on the meh side, or we put way too much focus on NPG/Cell/et al as markers for awesomeness.

  • Special Guest Lecturer September 22, 2015 at 7:08 pm

    Plus, Metchat had to have listed fake email addresses for each of her coauthors – otherwise they would have received an automated email from NPG when she submitted the manuscript. Even if this practice is not specifically disallowed by COPE, it is clearly unethical.

    I had always assumed that COPE guidelines would include that all authors approve the final version of a manuscript prior to submission. Is this really not the case?

    • Panagrellus September 23, 2015 at 3:00 am

      I think this is the case, but rather than collecting actual signatures (which would be a bit tedious for online journals), the submitting author usually has to confirm during the upload process that all authors approved the submission, and all authors should get a confirmation email.
      If those emails are fake, however…

  • genetics September 23, 2015 at 7:06 am

    I know of cases where good manuscripts stayed in the drawer of the PI for strange reasons. So I would not exclude that possibility, although at this point it’s just speculation on our part. But if that happens, there really is not much you can do about it in such a situation. And it can be extremely frustrating and career hurting.

    However, simply submitting the manuscript without consent of the co-authors is certainly no solution. Basically, that is fraudulent. An author can only be an author if he/she participates in writing the manuscript and personally agrees to the content. Claiming that this is the case while it is not is fraudulent and you deceive the readers.
    If this is not included in COPE guidelines, they should add it…..

    If at all, she could have submitted the manuscript as a single author and acknowledge the input of the others. At least that would have been a formally more correct submission, especially if she did most of the work. Certainly still a max confrontational move, and most likely not very helpful for the career either, but more difficult for the others to argue in favor of a direct retraction. It would have been a normal authorship dispute to be settled by the institution and both sides of the story would have been heard….

    • Conrad Seitz MD September 23, 2015 at 2:13 pm

      I agree with much of what you say. Clearly the author was frustrated by the failure of her supervisors to finish the process of creating a publishable document. If she could not get them to do the finalizing work or to approve her doing it, then perhaps she should have gone to the head of the department or the authorities at her institution. If she still couldn’t get satisfaction, then perhaps the boldest and most ethical thing to do would be to submit the paper as a single author and list the contributions of the others as so many papers do.
      If she has a legitimate grievance, it hurts her case to publish with fake email addresses for her co-authors: that looks like subterfuge and sounds unethical.
      I sense that she was unfairly treated by her supervisors and ignored; but that doesn’t justify publishing with fake email addresses. There must have been a better way, maybe not as fast but more ethical.

      • Federico D September 23, 2015 at 4:27 pm

        And the paper would have been retracted in the same amount of time, cause the former PI could have presented intellectual property issues at the journal.

        If it was (and I cannot say) a political issue (ie your former PI does not want to publish cause the topic is not “hot” enough at present moment) there is no solution that may work.

        • genetics September 24, 2015 at 8:47 am

          Well, intellectual property issues as well as authorship disputes (in the sense of proposedly missing authors) are something a journal cannot decide on alone. That must be handled by the institutions of the authors. You need an in depth investigation to clarify who generated the data, who really contributed etc.
          Lying to the journal by claiming that all authors agreed to submitting the manuscript while that is not true is something different. Especially if you use fake emails to hide the submission from your “co-authors”. And this obvious breach of a priori rules of the journal was the reason for the rapid retraction.

          By no means did I say submitting the manuscript as single author would have been a great and totally unproblematic idea……

          • Federico D September 26, 2015 at 4:15 pm

            Completely agree! I was not saying she used an equal solution thanwhat you were proposing: I was just saying that, in the view the PI does not want that paper out, submitting a single author would not be a solution, since if the work has been conducted in the PI institution and with PI funding, the intellectual property of the data is of the PI and PI institution. So the PI would have (maybe less) easily obtain the retraction.

            Again, what I mean is if the issue is political and not scientific, there is no solution

  • Frank Claire September 23, 2015 at 2:05 pm

    Its worth noting that in many labs, when someone leaves an unpublished manuscript behind and takes a new position, future students/postdocs may repackage the data and usurp first authorship. The motivation here may be to get their data published as first author before someone else poaches it.

  • AP September 23, 2015 at 3:21 pm

    Interesting situation. My first reaction: Was the paper reviewed and accepted? Is it solid, and there is nothing wrong with the data or conclusions? If yes, then there should be no retraction. The problem has to be resolved otherwise. Having information available which people can’t access over authorship disputes – that’s just silly. If rules have been broken, think of other ways of punishing the culprits. And as someone mentioned already, just the fact this story is out, is probably damaging enough for one’s career.
    I think this just points at the huge deficiency of the system. Not only of publishing, but the fact that one needs to wait for years for the information to come out. There are no arguments against alternative open science approaches, in which data is released in smaller blocks and published in near-real time. Such concepts exist and I hope will become reality one day. This sort of problems would simply never happen.

    • genetics September 24, 2015 at 9:15 am

      People have used the names of authoritative hot shots in their field as “co-authors” in order to increase the chances of getting the manuscript accepted. If someone would use your name that way, would you just shrug your shoulders and move on? You could argue that a correction is sufficient, but shouldn’t such an impertinence be sanctioned more severely, even if the work itself is not fraudulent?

      As to the publishing in near-real time. Well, I am not quite sure about that. Probably depends on the field, but there are so many ideas and data that turn out to be pretty useless. Trying to absorb all the ideas and raw data of every researcher in your field might not be that usefull after all. People complain about “salami publishing” and your “near-realtime publishing” would be salami to the tenth power. Personally, I do prefer to read full stories.

      • AP September 24, 2015 at 2:27 pm

        If this happened to me, I would certainly request my name to be removed from the publication. As I wrote above, whether or not there has to be another kind of sanction, depends on the case. But retraction is not the way.

        In my opinion, retracting already published research over a dispute between authors constitutes a violation of the Article 13 of the Charter of the Fundamental Rights of the EU (“The arts and scientific research shall be free of constraint.”). The postdoc would have every right to file a lawsuit with the European Court of Justice to enforce her rights and rights of all researchers and citizens of the EU to have access to that research.

        She can probably even make a case that she did what she did as a consequence of complying with the Article 13 (provided she can produce the evidence that the PI deliberately stalled the publication process and refused to publish within some reasonable amount of time).

        Regarding the system that can prevent this sort of thing, I didn’t mean to digress from the main story and go into details. But such concepts do not exclude full stories, and they provide technical solutions to filter relevant and important information. What I wanted to point out is that this problem can be easily prevented. Is it really true that Nature will put any author/email address on the paper without validating it? Wrong email address can be entered in a submission form for a number of reasons, e.g. due to a simple typing error. Sending a notification to that email address and waiting for a confirmation is a technically simple solution which would eliminate errors and prevent manipulation. In other words, it would be impossible for this event to occur. So why isn’t this implemented?

        • Narad September 24, 2015 at 5:45 pm

          Wrong email address can be entered in a submission form for a number of reasons, e.g. due to a simple typing error. Sending a notification to that email address and waiting for a confirmation is a technically simple solution which would eliminate errors and prevent manipulation. In other words, it would be impossible for this event to occur.

          Unless, of course, the person executing the fraud has the sense to have established the fake E-mail addresses in the first place.

          • AP September 24, 2015 at 6:45 pm

            Correct 🙂 A bit elaborate, but entirely possible. If the journal would accept only institutional email addresses, that would take care of it. This is also very easy to implement, but really hard to crack.

  • Special guest researcher September 24, 2015 at 8:57 am

    It is worth to talk about the behaviour of Ellenberg who has blocked the publication of her work ”ready to be published” since December 2013. Is is ethical? I always thought that the duty a group leader is to ensure the publication of postdocs and students working in his group. Is there any institution that can defend the right of students and postdocs in similar situation? I beleive that research in science needs more regulation and control. This will prevent any group leader to have the freedom to break the career of their postdocs in this way.

    • A crystallographer September 24, 2015 at 9:54 am

      We’ve only heard one side of the story, from someone who used fake contact addresses to prevent her unwitting co-authors from finding out she was publishing. I’m sure we’ve all seen enough stories where an over-enthusiastic PI pressured a junior scientist to write up dodgy results. How do we know this isn’t the opposite problem? If a PI is unwilling to publish (in Nature Communications! not Nature, or Science or Cell, but it still has “Nature” in the title!) even when the junior scientist has done all of the necessary work, what’s the more likely explanation: he’s doing it out of spite, or the work just isn’t good enough?

      I have no knowledge of this story or these individuals beyond what’s written above, and I’m not qualified to judge the paper itself, but I’ve never heard of a PI refusing to be associated with a paper unless there was something seriously wrong with it.

      • Federico D September 26, 2015 at 4:18 pm

        Sorry cristallographer, plenty of situations in which big name PI are not publishing anything less than Nature, Science or Cell, and if the story does not fit the level, they prefer to close the story in the desk drawer.

  • Toby September 24, 2015 at 10:22 am

    Publishing in Nat Commun is very expensive (clearly Commun does not stand for what some might think it does), something like 5,000 US. Who forked out? Is it refunduble?

  • Ha September 26, 2015 at 3:39 pm

    A Nawn
    But if a Nat Comm. is pivotal to getting those job talks…then either the rest of her record is more on the meh side, or we put way too much focus on NPG/Cell/et al as markers for awesomeness.

    If you are in science, you already know which one it is…

  • Post a comment

    Threaded commenting powered by interconnect/it code.