Research integrity

The stinking fish of Strasbourg

A 2014 publication from the lab of Laurence Maréchal-Drouard, director of CNRS IBMP institute in Strasbourg, seems to contain evidence of inappropriate data manipulation. Other versions of this problematic figure appear in 2 PhD dissertations supervised by Drouard

There is a popular saying: “A fish stinks from the head down”. This is what the situation starts to look like for French CNRS Institut de Biologie Moléculaire des Plantes (IBMP) in Strasbourg, also unofficially known as “The Olivier Voinnet Institute for Research Integrity in Plant Sciences”, after its most famous former research director. A publication from the lab of the institute’s head, Laurence Maréchal-Drouard, seems to contain evidence of inappropriate data manipulation. The gel shown in Figure 1A of Salinas et al 2014, seems to have “no less than 4 copies of a background patch”, according to a concerned reader who contacted me with this evidence. Drouard herself did not reply to my email, possibly because she and her entire CNRS institute are banned from communicating with me.

The IBMP director Laurence Drouard as now part of data integrity tumult herself. The background in this published gel lane from her lab seems to be duplicated four times.

A tenured scientist in the Droaurd lab, Thalia Salinas-Giegé, is the first author on that publication. Her husband is Philippe Giegé, and a paper from his own lab proved to contain duplicated gels (Gutmann et al 2012, PubPeer evidence and Giegé’s response here). As aside, it was Giegé’s tenured researcher Géraldine Bonnard who also issued the communication ban with me to her IBMP colleagues. The middle one of the three co-authors on the problematic Gutmann et al 2012 paper is Anthony Gobert, who is married to Hélène Zuber, a tenured member of a rather problematic IBMP lab of Dominique Gagliardi. Why is this lab problematic? Because Gagliardi’s other tenured lab member, Heike Lange, is a person with a record one should be most cautious with. This German researcher has been accused of western blot band duplications in her earlier papers from the lab of Roland Lill at the University of Marburg (Lange et al 2000 and Lange et al 2001). With Gagliardi, Lange published a paper for which she recently had to admit more band duplications on PubPeer, for Lange et al 2011. She wrote there:

“I regret making the wrong control panel in the first place but it was never made to be published instead of the original control. Its insertion in the figure was a genuine mistake. Please note that both the creation and the publication of this panel was a mistake of me, Heike Lange, and none of the other authors knew of its existence. These issues are being communicated to the journal editor”.

Evidence of band duplication in Lange et al 2011 paper from Gagliardi lab, as posted on PubPeer.

Just some days ago, IBMP had to retract a paper from the lab of Véronique Ziegler-Graff for data manipulation in four figures (Pfeffer et al 2002). A co-author on that paper was Voinnet’s former right-hand man, self-admitted data manipulator and currently IBMP group leader, Patrice Dunoyer.

Given all this data manipulation chaos, it is rather fitting that this institute will be visited tomorrow, on March 8th by a travelling circus of research integrity evaluators, also known as “Comité HCERES” scientific review board” (see my report). The members of this evaluation committee themselves had to admit data manipulations in their own papers. Alain Tissier and Ute Vothknecht sent their first authors to PubPeer to admit data manipulations for Rontein et al 2008 and Chigri et al 2006, respectively. Martin Crespi and his first author admitted data manipulation on PubPeer for Campalans et al 2004. Everyone however  dismissed own data manipulations it as scientifically irrelevant. A previously corrected paper from Crespi lab (Elvira-Matelot et al, 2016) had a duplicated gel band introduced by his postdoc, as the paper’s other corresponding author, Hervé Vaucheret, told me:

“I’d like to point that the author responsible for the image manipulation in Elvira-Matelot et al, 2016 that was recently corrected is Florian Bardou. […] Thus, this manipulation was not done in my lab. Was I aware of this manipulation? NO ! Martin Crespi, the supervisor of Florian Bardiou also was not aware of this manipulation”.

In addition to other IBMP scientists whom I already listed in my earlier report,  also the tenured IBMP researcher Thierry Heitz, member of Danièle Werck lab, had concerns about the data integrity in his publications flagged on PubPeer. Here is an example flagged on PubPeer for La Camera et al 2009, where Heitz is last author: 2017-03-07 12-36-06 heinz

Tomorrow’s HCERES evaluation will therefore be a great fun for sure. One does wonder if the IBMP or CNRS in general learned anything at all from the Olivier Voinnet scandal, except of the importance of a thorough cover-up and a communication ban. The CNRS currently have another fresh data integrity scandal on their hands, with yet another plant scientist: Susana Rivas.



If you are interested to support my work, you can leave here a small tip of $5. Or several of small tips, just increase the amount as you like (2x=€10; 5x=€25). Your generous patronage of my journalism will be most appreciated!


12 comments on “The stinking fish of Strasbourg

  1. A new minor issue (lane splicing) has been highlighted in this Pubpeer thread.

    Not a big deal really, but the interesting point is that this article is from 2016! By that time everyone was more than aware of gel image manipulations issues in her lab. Still, some seem just immune to any ethical concerns.


  2. Habits die hard and if you have been trained to splice and not trained in the importance of data, then it is difficult to change.


  3. Pingback: CNRS lab of Voinnet’s right hand Patrice Dunoyer dissolved by director’s decree – For Better Science

  4. Pingback: Voinnet’s sidekick Dunoyer welcomed at Nature Plants, despite retractions and admitted misconduct – For Better Science

  5. Pingback: Voinnet’s CNRS investigator Catherine Jessus with own data integrity issues – For Better Science

  6. Pingback: John Innes Centre director Dale Sanders rebuffs a non-peer reviewed internet troll – For Better Science

  7. Pingback: CNRS hits back at the stream of misconduct evidence – For Better Science

  8. Pingback: Call to research integrity, or at least a minor revolution at CNRS – For Better Science

  9. As explained here

    the Salinas et al. 2014 NAR article has some more interesting stuff in supplemental figure 3. The authors show two obviously wrong models that do no fit data in the same plots. Their comments about the dissociation constants should be reevaluated (VDAC36 Kd 1.4x higher than VDAC34 Kd).
    It is strange that this issue was not caught during initial peer review in 2014, or more recently in the review that must have taken place on the occasion of correction of a doctored images in the same paper.
    I would like to hear from NAR about the process of article correction in this specific case.


  10. I had a closer look at the corrigendum published for the Salinas et al. 2014 NAR article. In this corrigendum, only figure 1A is acknowledged as unethically manipulated. In contrast, Suppl. Figure S4B is only said to have been reorganized using undisclosed splicing to show lanes in increasing size order. This is wrong, as the D5 lane in the paper shows portions of lane D6 were used to obscure unspecific signal. The D3 lane in the paper is nowhere to be found is the raw data presented in the correction. Indeed, the corrected figure D3 lane matches the new raw data, but not the original D3 lane in any way. The D7 NW lane in the paper also doesn’t match the raw data presented in the correction. I’m surprised and disappointed that corrections are not better investigated at @NAR_Open.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: