A team of researchers in France has lost two papers on their studies of yeast because the work was “a complete work of fiction,” in the words of one colleague.
The papers came from the lab of Jean-Luc Parrou, of the University of Toulouse, and involved work by a former PhD student of his named Marjorie Petitjean. According to Parrou, institutional investigators are now in the process of revoking her degree, but have been delayed by the Covid-19 pandemic. We could not find contact information for Petitjean.
One of the articles, “Yeast tolerance to various stresses relies on the trehalose-6P synthase (Tps1) protein, not on trehalose,” appeared last year in the Journal of Biological Chemistry. It has been cited 54 times, according to Clarivate Analytics’ Web of Science. The other, “A new function for the yeast Trehalose-6P Synthase (Tps1) Protein, as key pro-survival factor during growth, chronological ageing, and apoptotic stress,” was published in Mechanisms of Ageing and Development in 2017. It has been cited 11 times.
The retraction notice from JBC reads:
This article has been withdrawn by Marie-Ange Teste, Jean M. François, and Jean-Luc Parrou. Marjorie Petitjean could not be reached. The corresponding author identified major issues and brought them to the attention of the Journal. These issues span significant errors in the Materials and Methods section of the article and major flaws in cytometry data analysis to data fabrication on the part of one of the authors. Given these errors, the withdrawing authors state that the only responsible course of action would be to withdraw the article to respect scientific integrity and maintain the standards and rigor of literature from the withdrawing authors’ group as well as the Journal. The withdrawing authors sincerely apologize to the readers and editors.
The one for MAD states:
This article has been retracted at the request of Marie-Ange Teste, Isabelle Léger-Silvestre, Jean M François and Jean-Luc Parrou. Marjorie Petitjean could not be reached.
The corresponding author identified major issues, brought them to the attention of the Journal.
These issues span from significant errors in the Material and Methods section of the article and major flaws in cytometry data analysis to data fabrication on the part of one of the authors.
Given these errors, the retracting authors state that the only responsible course of action would be to retract the article, to respect scientific integrity and maintain the standards and rigor of literature from the retracting authors’ group as well as the Journal.
The retracting authors sincerely apologize to the readers and editors.
Parrou told us that he first noticed serious problems with Petitjean’s data in 2016 when he was preparing a manuscript for submission to Science. At the time, his former student was working as a postdoc at a lab in Scotland:
Since that time, followed dreadful discoveries of misconducts and erroneous data no longer reproducible.
This nightmare period therefore lasted for almost 4 years (4 years next summer), during which I also tried to get answers from Marjorie Petitjean about the reasons and motivations for such a misconduct. With no success.
Parrou said he informed research integrity officers at several bodies, including his university and national institutions. The National Institute of Applied Sciences (INSA), which hosts Parrou’s lab and which awarded her PhD, launched an inquiry into Petitjean, he said:
Two nominated experts hence came to the lab last autumn and provided their own conclusion on that story. They obviously confirmed what I discovered. With these elements in hands (my report and this expertise) , the INSA decided to open a disciplinary procedure, which started last February, to formally revoke Ms. Petitjean Ph.D.
Parrou said that Petitjean declined to defend herself during the investigation or appear before the committee, which was scheduled to render its final decision in March but has postponed the ruling.
Meanwhile, Parrou said he learned that Petitjean’s post-doctoral supervisor found evidence that she had been fabricating data in that position.
‘A certain feeling of guilt’
For Parrou, the experience has been wrenching:
Today, my greatest regret is not having an answer as to the motivation of this student. … But as a thesis supervisor, this will never take away a certain feeling of guilt.
Of course, this story led me going through a period of great doubts, less and less as the time was going on, and finally no doubts at all with those post-doc revelations ; therefore, time might have been saved going quicker through that nightmare, but an inadequate decision from my part towards Mrs Petitjean, with unfounded or partially wrong accusations, could have been dramatic, also for her.
It is a very difficult task to start, from personal initiative, such a complex investigation work with the very probable objective to withdraw articles, in which you invested money, energy and enthusiasm while doing the job, discussing and writing. While you planned great perspectives. But the honesty must be the rule in science, for pairs and citizens’ trust in science. I had no choice. And fortunately got the support from my colleagues and co-authors, as well as from research institutes.
Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on Twitter, like us on Facebook, 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].
Second paragraph: “Peitjean” instead of “Petitjean”.
Fixed, thanks.
Exactly
“Doubts” … doubts so tenacious. Despite a very fastidious work trying to clarify that issue since the summer 2016, and increasing evidence of misconduct by this person, I left her the benefit of the doubt. And I likely denied myself accepting that it was a (conscious) will to cheat and produce falsified data ; I was just trying to find out what was true in published data. With the revelations from her supervisor in post-doc, my residual doubts have been fully dispelled.
No more doubts, but the damage has been done. A nightmare vision, coming to darken years of research spent imposing myself and transmitting rules of rigor and honesty in science. To withdraw the articles … or ‘wait and see’ ? Actually, one single option : the first one. Careers broken, including hers, but the feeling to leave sound science to my pairs.
I hear you about “doubts”. I had a research student who never seemed to be able to put my feedback from our weekly office meetings into practice. It all fell into place one day when she emailed a draft of her thesis with the “track changes” comments still in situ.
One of the comments was from someone I had never heard of. It was a question to the student along the lines of, “Is this what your supervisor wanted?”
It suddenly occurred to me that the student was getting someone to ghostwrite her thesis, which explained why the feedback I was giving to her in person was never applied.
I then became worried that her data spreadsheet might not be based on real data. I asked her, repeatedly, to bring her raw data. She refused, then requested and obtained a new research supervisor, and graduated soon after.
I was later informed by the Faculty that I had been included as a co-author on a published paper in recognition of the time spent supervising this student. So I am left with co-authorship of a paper I have doubts about, but no hard evidence either way.
Jean Luc, roughly what percentage of the data was fabricated?
Well, I think part of the problem is that being a grad student or post doc in an academic institution is just too crappy of a job (bad pay, no job security), and some people just try to cheat to get themselves out of it into a better paying and respectable job.
Maybe the system needs to be fixed, and that starts with faculty and administrators. I have a little sympathy for faculty who have been fooled by cheaters, but not too much. Inevitably, the academic system leadership…the managers of the data generators…expects from the data generators way too much and, at the same time, refuses to pay and give them the respect they deserve.
Having self-funded a PhD in a discipline with very limited funding, I find this explanation equal parts ridiculous and worrisome. Getting a PhD is not something you have to do. We need to better support grad students because their work in the backbone of research and academia. Yes, funding is terrible and grad student pay is laughable but if your real goal is to become a higher paid academic you aren’t in it for the right reasons. But you can’t write off data fabrication and manipulation as a means unto an end. If you don’t care about the integrity of your PhD research you don’t care about advancing knowledge—only advancing yourself.
Well, I mention possible reasons, but I don’t think they are good ones; the ends (good job in academia or industry) does not justify the means (cheating to get out of a crappy job).
I’m suggesting reasons as to why cheating (faking data) occurs. The incentive structure (to get a great job) needs to be changed, and I suggest lowering of academic tenure track scientist pay and increasing pay of post docs and grad students so there is less of an incentive to cheat, as the job in the end would not be that much better.
If you don’t want to ask faculty to do their job, namely, watch over primary data generation (which is what they should be doing), then there needs to be a compensatory reduction of pay for the faculty member. Almost all faculty behave this way, so there needs to be universal reduction in faculty pay.
Instead, faculty are coddled with great pay and a job they can never lose. That is a great incentive; I would be tempted to cheat too if I didn’t have a good moral sense established by my parents and culture.
Message from INSA Toulouse.
The disciplinary procedure concerning Ms. Petitjean is still ongoing. As long as the judgment has not been pronounced, the INSA can neither confirm nor deny the conclusions reported in this post.
Message from INSA Toulouse.
The disciplinary section of INSA Toulouse, held on July 1, 2020, has judged from an independent expertise and the hearing of Ms. Petitjean, that she committed breaches of scientific integrity during her PhD thesis. Therefore, Ms. Petitjean doctoral diploma will be withdrawn.
I faced a similar situation as a doctoral student while following up on someone else’s PhD work that preceded my PhD program (with partial overlap). I noticed several anomalies in this work in 2016 and have since been intermittently following up on it. It is not easy to follow through especially when the supervisors of the concerned work are not very keen to address it (they were my doctoral supervisors too and I immediately reported my findings to them). Kudos to Jean-Luc Parrou for following up on it all through these years.
It is not easy to tolerate such practices when you are involved in related work. I am a co-author in a couple of publications in connection to the work I have evoked. After failing to get an action from the concerned authorities so far, I have posted my concerns on PubPeer (my post as a co-author of a publication: https://pubpeer.com/publications/A4AA4DAC550DE014EC874D3EAC95F0#11). At present, I am trying to get the higher authorities at the university to address this matter, but this story indicates that this might drag on for a few more years. Sadly, there is some indication that some questionable practices do not stop at the work I followed up during my PhD.