Retraction Watch

Tracking retractions as a window into the scientific process

Neuroscientist who threatened to sue Science-Fraud.org, retracted two papers is out at Tufts

with 57 comments

Domnez_Gizem

Gizem Donmez, via Tufts

Gizem Donmez, a neuroscientist who has retracted two papers from Cell and the Journal of Biological Chemistry, is no longer in her position at Tufts University, Retraction Watch has learned.

A Tufts spokesperson confirmed the news for us yesterday:

Dr. Donmez is no longer employed by Tufts University. We don’t, as a rule, comment on employee personnel matters.

We’ve tried to contact Donmez for comment, and will update with anything we learn.

Written by Ivan Oransky

October 16th, 2014 at 9:00 am

Comments
  • AnonyMoose October 16, 2014 at 10:06 am

    Science-Fraud.org seems to be slightly empty.

  • ferniglab October 16, 2014 at 10:19 am

    Rather late for the community to say “I told you so”, the damage of “lawyering up” by people who do have something to hide is done and cannot be undone.

    • aceil October 16, 2014 at 12:39 pm

      ferniglab,

      Wonder if it can be recovered!

    • Brad Casali October 16, 2014 at 8:01 pm

      Everyone has a right to defend themselves, regardless of the intention and/or whether or not they are guilty of hiding something. In my opinion, science fraud was a clearly defamatory site (look at the name), and the (at one point nameless) creator also ‘lawyered’ up when he was threatened to be sued.

      In my opinion, you can’t post material on a site called ‘science fraud’ and not assume that there was no intention for it NOT to be defamatory and potentially damaging to a career (especially if it’s not true).

      • Neuroskeptic (@Neuro_Skeptic) October 17, 2014 at 6:29 am

        But was anything on there not true?

        And it’s not “lawyering up” if someone else starts it!

        • Brad Casali October 17, 2014 at 11:08 am

          You missed the point. It doesn’t matter what the truth is, my point was that everyone has a right to defend themselves, regardless of the truth of the matter.

          For what it’s worth, it is rather obvious that there are problems with her papers. But I don’t think she should be faulted for defending herself.

          If it happened to you (and let’s say that you didn’t clearly fabricate images), then you’d do the same thing (defend yourself). Heck, even if you did, you’d defend yourself.

          • Neuroskeptic (@Neuro_Skeptic) October 17, 2014 at 12:04 pm

            No, if you knowingly do something wrong and then defend yourself – against charges that you know to be true – then you are committing misconduct all over again.

          • QAQ October 17, 2014 at 6:50 pm

            So defending oneself against a charge that is true is prima facie wrong?

            What if the punishment is excessive? Would you say it would have been wrong of her to hire a lawyer if her crime was having kissed a consenting adult and the punishment was death? Probably not.

            Fraud implies a deliberate practice and is the name of the website. I do not believe that concrete evidence of fraud by her (including concrete evidence of her deliberate practice) was made available. Therefore, I don’t see how you can be so condemning of what she did. Perhaps she feels that her crime was in fact, not fraud but laziness and that it would be been more fair to speak of her on a site called science-messups.com.

            She’s probably wrong here, but if she legitimately thinks that she didn’t commit fraud OR that the punishment doesn’t fit the crime, I don’t see why her hiring a lawyer is such an awful thing.

          • Neuroskeptic (@Neuro_Skeptic) October 18, 2014 at 6:51 am

            “If the punishment is excessive” then maybe but that’s not what we’re discussing here, the issue is whether she should have threatened to sue Paul Brookes for exposing her misconduct.

            If she honestly believes she did nothing wrong – then to be honest she has no place in science anyway.

          • QAQ October 18, 2014 at 4:01 pm

            Did she threaten to sue Paul Brookes for exposing her misconduct OR did she threaten to sue him because she believed that he mischaracterized her misconduct?

            Not all misconduct is fraud. This is extremely evident from the definition of the two words. Misconduct may include negligence, while fraud requires conscious intent. A reasonable person may think that Brooks, through the title of his blog, was implying that she committed the later.

            Being that you are completely unable to look into her mind, I think that it’s extremely unfair to make judgments about her intent on any matter.

            In science, as in law, one is taught the extreme importance of word choice. Personally, I think that it’s a bit unfortunate that Brooks chose to name his blog with fraud and then discuss cases that were not fraud beyond a doubt. I don’t expect to find immunology articles in a cardiology journal.

          • Narad October 17, 2014 at 9:08 pm

            You missed the point. It doesn’t matter what the truth is….

            It very much does if the matter at hand is defamation, of which you made a blanket assertion. Everyone in the U.S. is entitled to sue, but this is not necessarily the same thing as the Platonic notion of “defending oneself” that you seem to be invoking.

          • Brad Casali October 18, 2014 at 7:09 pm

            I doubt her going to Science Fraud and defending herself would do any good (hence why I equate defending herself to bring suit). Just look at some comments on PubPeer. The AUTHORS produce actual images that are in question, and the ANONYMOUS ‘peers’ say either the image is not high resolution enough or it’s not good enough.

            Though, some are satisfied with the authors’ responses, there are about the same quantity of posters who aren’t. Hence, you bring in the lawyers.

      • Scotus October 17, 2014 at 9:30 am

        I think that’s a little bit unfair. Data and conclusions can all be defended by publication, replication and advancement of the research.
        Paul’s only possible failing with Science Fraud was that some of the commentary strayed beyond factual into opinion and was often accompanied by a liberal dose of sarcasm. Which was of course why I enjoyed reading his postings.

        • Brad Casali October 17, 2014 at 11:11 am

          And the only possible failing part of your comment was the problem with that site. People who are anonymous on the web will troll out all sorts of things, mainly because there is no consequence for their actions. In that case, he clearly did suffer consequences.

          But what do you expect by creating a site called, “Science Fraud?”

          • Neuroskeptic (@Neuro_Skeptic) October 17, 2014 at 12:05 pm

            I’d expect it to feature cases of science fraud, which it did!

          • Scotus October 18, 2014 at 6:44 pm

            Yes, I agree that the title of the site was inflammatory. Yes, I agree that some individuals use their pseudo anonymity to attack others unfairly. I think, like a lot of us, Paul Brookes is an honest toiler who just got so fed up with other people getting on by cheating that he just had to do something about it. And, as I recall, he works on mitochondrial biology so the sirtuin stuff would have been right in his wheel house. Maybe he was actually interested in reproducing some of the work?
            I am sure she ended up wasting a lot of money on the attorney but perhaps this bought her some time to plan an exit strategy or allowed her to ask Tufts what might happen if they had an investigation and found her responsible for the alleged misconduct. Or perhaps she really thought she was being falsely accused. Since a finding of guilty results in the scientific equivalent of the death penalty why not try everything possible if you know you are going down in the end?
            Maybe if/when these cases become more commonplace lawyering up for purposes of plea bargaining or slowing down the process might be a viable strategy to mitigate the outcome. Its not as if these things don’t take a lot of time and money to resolve.

          • Brad Casali October 18, 2014 at 7:18 pm

            I have nothing against Paul Brookes. I have no doubt he is a meticulous and dedicated scientist and was no doubt fed up with people cheating to get ahead. The only major concern I have is HOW he went about bringing the cheating of others to light.

            I personally have bone to pick with people who post anonymous accusations online. It makes me very angry when others hide behind anonymity.

            And YES, I am posting these comments under my real name.

          • Narad October 18, 2014 at 10:15 pm

            The only major concern I have is HOW he went about bringing the cheating of others to light.

            You have flatly stated that the name of the site was prima facie evidence that the whole of its content was defamatory to the standard that would apply to a public figure:

            In my opinion, you can’t post material on a site called ‘science fraud’ and not assume that there was no intention for it NOT to be defamatory and potentially damaging to a career (especially if it’s not true).

            That said,

            I personally have bone to pick with people who post anonymous accusations online. It makes me very angry when others hide behind anonymity.

            Anger about “anonymity” (I suspect that this may include pseudonymity) does not seem to be a sound footing for evaluating the content of a writing. Are you similarly incensed by the Federalist Papers?

            And YES, I am posting these comments under my real name.

            Do you think this somehow presumptively grants them a lowered standard of scrutiny?

          • Brad Casali October 19, 2014 at 10:46 am

            To your last point: No.

            To your other points: I have no idea what you’re trying to get at. This is my opinion, and you can take it or leave it. If you have a problem with it, go blog on the interweb under a pseudonym.

          • Neuroskeptic (@Neuro_Skeptic) October 19, 2014 at 11:32 am

            If you dislike anonymous comments then you have a right to ignore them.

            Everyone has that right.

            You should be getting angry at the (rather high) percentage of people who choose to take anonymous comments seriously. Not at the anonymous comments themselves.

  • Leonid Schneider October 16, 2014 at 2:36 pm

    Call it Schadenfreude, but I am happy there is some justice left. Luckily Dönmez was not tenured or senior enough, which made this separation easier for Tufts.

  • Scrutineer October 16, 2014 at 3:26 pm

    It can’t have been easy for the Tufts admin to see their star hire revealed for something other than they expected.

    And as documented here,

    http://www.psblab.org/?p=167

    one might infer that Cell Press quailed at the thought of having to retract a star paper from a star lab at the star institute that just happens to own the building they work in. Shit happens; but in the end the paper had to go. (289 citations according to Google Scholar: pretty nifty for a 2010 paper.)

    For what it is worth, it is almost the 10 year anniversary of the first of the papers authored by the good doctor (during her European sojourn) which have been noted to deploy those trademark images in the figures.

    https://pubpeer.com/publications/799A8855052AC9153D14F1CB1BA6BC

    Luckily the affected journal RNA is not domiciled in Boston so surely does not suffer from the editorial dissonance that Cell Press might have suffered from. But only time will tell as to whether it has editors that can recognise lane duplications when they are helpfully marked out with red arrows. Readers here will know by now that many journals apparently rely upon editors not at all equipped with such skills. But if you are new to these pages, you could follow the first link above to see which category Journal of Neuroscience belongs to.

  • Leonid Schneider October 16, 2014 at 3:51 pm

    I have a business idea: there are services for ghost-writing of PhD theses ( I actually saw their job recruitment ads, in Germany!), so why not offer a discreet and reliable service for faking data by IT professionals, so that stellar careers may not be endangered again by amateurish image duplications and photoshopping? Nudge Nudge

    • Dave Langers October 17, 2014 at 6:45 am

      Yes, and then have some Edward Snowden figure among their ranks who opts out after a few years, taking all client records with him and exposing all “scientists” who couldn’t resist the seduction of taking this fraudulent path. That might be a neat cleansing operation.
      (Let’s first build a system where all exposed fraudsters have to properly fear for their life-long careers though… nudge nudge)

    • BB October 17, 2014 at 1:26 pm

      Splendid idea, I am certain that it would generate lavish revenue. We sometimes also use to do some brainstorming with my colleagues about a fictional software (VirtuaBlot) the first ever western blot simulator, that would be able to render realistic, ready to use gel images and supporting data using your already written results section as the only input. This software would be able to save high-profile “researchers” with a tendency towards creative data management from the pestering nitpickers of pubpeer. No more shady band lane duplications, or embarrassing splices! VirtuaBlot will keep you safe in the turbulent world of PPPR!
      (beta-testeers wanted!)

  • Leonid Schneider October 16, 2014 at 4:01 pm

    This is a bit unsettling, whatever this is: http://gizemdonmez.wordpress.com/ and http://gizemdonmez.blog.com/
    Not many people take to praising themselves in third person….

    • akrasiographer October 16, 2014 at 4:20 pm

      These articles are obviously some sort of attempt, probably using semi-automated software, to clean up her internet image.
      Also, when googling her name in Europe, “Some results may have been removed under data protection law in Europe.”
      The number 1 hit is still an article by Paul S. Brookes, so she doesn’t seem to be succeeding.

      • Leonid Schneider October 16, 2014 at 4:32 pm

        The most recent posts were from September 2013, shortly after JBC retraction (manageable) and almost a year before the Cell retraction (deadly). So what was the point of this and who wrote it, if not Dönmez herself?

        • DrDoo October 16, 2014 at 5:37 pm

          A similar strategy was used by Anil Potti to push unflattering search engine hits down on the list of results.

          • Scrutineer October 17, 2014 at 1:43 pm

            Well remembered DrDoo!

            The best part though is that she used Potti’s reputation-enhancing firm to create her website. Sometimes you can’t make this stuff up.

          • DrDoo October 17, 2014 at 4:13 pm

            I didn’t catch that!
            I wonder if that firm’s business strategy is to solicit folks with PR troubles.

          • JATdS October 19, 2014 at 4:44 pm

            Good catch, Leonid. One of the sites states “Gizem Donmez is not native to Massachusetts or the United States. Originally, Gizem Donmez started her career as a student in Turkey. She only spent her undergraduate years in Turkey before completing her master’s and PhD in molecular biology in Germany. However, as a scientist in Massachusetts, Gizem Donmez was invited to participate in the Science and Engineering Diaspora Engagement in February of 2013. The engagement was held by NODES, which is a partnership between the U.S. Department of State, the American Association for the Advancement of Science, the National Academy of Sciences, and the National Academy of Engineering.” That might be saying something. On the other hand, if we can prove that these sites have been manipulatively written by Donmez herself, could there be some action to get them removed by Gogle, and how does the new laws in the EU that regulate the removal of private information work in this case?

          • Leonid Schneider October 19, 2014 at 5:47 pm

            Why removing it? It does look embarrassing, retrospectively, but so it should. Let all read what an exceptionally gifted scientist Dönmez was celebrated to be, now that we know how her scientific fame was made 😀

          • Narad October 19, 2014 at 7:58 pm

            how does the new laws in the EU that regulate the removal of private information work in this case?

            There is no new law; there was a decision (PDF) regarding an old one. There is no provision for third-party removal of “manipulatively written” self-promotion.

          • DocMartytn October 21, 2014 at 9:09 pm

            Her thesis is online and looks rather good; the quality of the gel images ‘improved’ as her career advanced, initially wavy and sloping, but later pristine.

      • Narad October 16, 2014 at 11:53 pm

        These articles are obviously some sort of attempt, probably using semi-automated software, to clean up her internet image.

        This is about the worst attempt at reverse SEO imaginable. Check out the link to “follr,” which leads to (“Website”) a link to “ibosocial.”

      • herr doktor bimler October 17, 2014 at 4:31 pm

        The literary antecedents for reputation enhancers are not flattering:
        http://en.wikipedia.org/wiki/The_Repairer_of_Reputations

        • Narad October 17, 2014 at 9:17 pm

          Having seen the applicant-screening test for one prominent U.S. purveyor of such services,* I can state that the client offering at the introductory tier didn’t look promising.

          * Perhaps recognizable to locals for sponsorship announcements on nationally distributed public-radio programs.

    • ED October 19, 2014 at 10:22 pm

      For the sake of lab happy hours everywhere, I hope someone archives these brilliant jewels.

  • Leonid Schneider October 16, 2014 at 4:46 pm

    This Dönmez paper in Mol Cell is so far uncommented on PubPeer: https://pubpeer.com/publications/B3DF7D8791ABE271570E77B729FA67
    Tech note: Ifor some reason it does not come up through her name search, only by doi.

  • herr doktor bimler October 16, 2014 at 7:10 pm

    Her 2004 exercise in collage appears to remain unretracted:
    https://pubpeer.com/publications/799A8855052AC9153D14F1CB1BA6BC

    • Stewart October 17, 2014 at 3:38 am

      So, the accusations of potential image manipulation on the science-fraud website were correct?

      Pubpeer has an interesting image analysis link on one of the gomez papers:
      https://pubpeer.com/publications/799A8855052AC9153D14F1CB1BA6BC
      http://i.imgur.com/JNF4ngN.jpg

      Dr Brookes: http://www.psblab.org/?p=167
      One editor replied to Dr Brookes concerns regarding the potential science fraud “institutions produced detailed reports that included the original images used to construct the figures” and “there was no evidence that any results had been misrepresented” “We consider the matter closed”

      That does appear to be alot of hard work by Dr Brookes, to seek the truth on work that is now retracted/under scrutiny by the science community.

  • Scotus October 17, 2014 at 9:06 am

    Interesting. I assume that, like at our place, as an Assistant Prof her appointment was probationary and reviewed on some sort of a recurrent basis and they just decided not to reappoint her. Much easier than dealing with someone who has tenure. However, I bet she turns up somewhere else, probably back in Europe where any potential eventual NIH sanctions won’t matter.

  • DocMartytn October 18, 2014 at 4:28 pm

    She got the job, and another researcher didn’t, because her papers got into Cell and JBC. They got into Cell and JBC because they were pretty and had a clear A->B->C->D narrative. If you have non-pretty data and A->Bish with a bit C, but then D, then you are rejected without going to peer review.

    I don’t trust the vast majority of Western blots published, as I have run a few.

    I was interviewing a candidate a month ago and he had run iNOS gels in cancer cells and showed a single band at 125 KD. I said that I always found more post-translationally modified iNOS than canonical, with truncated and possibly SUMOylated-iNOS.
    He told me that he got rid of unwanted bands in the Gels in his pre-print by dropping the secondary antibody level until he got the band he wanted. He thought that was more ethical than photoshop.

    • Leonid Schneider October 19, 2014 at 6:11 am

      DocMartytn, what you give as an example for cheating, is in my view not cheating at all, but academic disagreement. You all working in the NOS field can discuss which band is more important and what the correct protocol for WB may be. What you applicant did, was trying to develop his own working protocol for iNOS detection, you cannot accuse him of fraud similar to photoshopping. This would be crudely unjust.

      • DocMartytn October 19, 2014 at 10:02 am

        You know, and I know, that when you examine GFAP in astrocytes one gets a least three species; the ratio of which can vary depending on what you have hammered them with. One can show the region between 40 and 70KD, with all the bands, or just show the 50KD band. We both know that the 50KD band is not always the major band. You Leonid, show the region, warts and all, but this is by no means universal is it?
        If you observe a transition from 50KD to 70KD, but omit to include the upper portion of the gel, it is deliberate misrepresentation of ones data.

        Have a look at Figure 2

        http://www.nrronline.org/article.asp?issn=1673-5374;year=2014;volume=9;issue=4;spage=385;epage=393;aulast=Zhou

        • Leonid Schneider October 19, 2014 at 12:49 pm

          No DocMartytn, it is not. I will not debate your issues with (my?) publications on GFAP here, feel free to contact the journals if you need to. btw, the link you posted shows in Fig 2 an amateurishly 3-lane-spliced WB for GFAP, which is problematic indeed.
          There may be different views on which protein band is an important splice isoform or post-translationally modified, but what it boils down to, is this: you could be right today and wrong tomorrow when new experimental knowledge arrives. Maybe even the poor interviewee was right after all, and not you, this all is academic. You both will probably be able to cite papers to support your diverging views.
          But faking a desired result using photoshop or image duplication/modification as Dönmez did, always was, is and hopefully will be “well below the standard that we expect,” in Donmez’s words.
          Do kindly not mix these things up, and please refrain from imposing such doctrines on subordinates and students.

          • DocMartytn October 19, 2014 at 2:09 pm

            I was actually praising your work; you present the actual gels, using wide enough slices to show multiple isoforms of proteins. The fact that you have distortions from the horizontal makes publishing these features in ‘top’ journals very difficult; not pretty enough.
            However, knowingly dropping reporter or under-developing films so as to only allow visualization of the canonical band, is IMO just as bad as photoshoping them away.

          • Leonid Schneider October 19, 2014 at 4:25 pm

            First, under-developing films is only unacceptable when you pretend to show a knock-out, while you still have some background expression. Otherwise, showing the band you assume (from literature!) is the correct one while ignoring others, may reflect poor scientific quality or lack of knowledge in the matter. But again, this is not an intentional fraud. This would be, in the worst case, incompetence or “honest error”. Why you still insist this is just as bad as deliberately inventing data, is beyond my understanding.
            Actually, in my experience, by “hiding” bands in the matter you describe as fraudulent, one tends to overlook important results which could have made the paper even stronger.

        • JATdS October 19, 2014 at 2:51 pm

          Interesting, related to the Zhou et al. NRR paper you reference, how one author is an author for the followng function alone: “Wang H revised the manuscript.”

    • Brad Casali October 19, 2014 at 10:53 am

      Maybe you didn’t quite want the candidness? Some scientists do things a bit ‘differently’ and in my experience, I often don’t want to hear about how the lengths they went to get results.

      And about your journal idea: I’m sure her working for a big wig helped quite a bit. You know, “Cell” has a very ‘rigorous’ method of allowing those papers through to peer review whose author’s PI is famous and has published there before.

      Just my 0.02.

  • genetics October 21, 2014 at 5:15 am

    Just a side note: When does a retraction appear in pubmed and in PMC?

    The retraction of the cell paper is now more than two months old. Yet, you still don’t see it in pubmed: http://www.ncbi.nlm.nih.gov/pubmed/20655472 . In PMC you can still find the entire paper without any indication of retraction: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2911635/

    That is a painstakingly slow correction of the literature…..

  • Post a comment

    Threaded commenting powered by interconnect/it code.