The revolution in French science has begun, and its target is the corrupt elite who betrayed against basic principles of research integrity and conspired to protect one of their own, the CNRS chief biologist Catherine Jessus. French scientists seem to finally have enough of watching fraudsters eat cake and now storm the Bastille.
Much of evidence of data manipulation in papers from Jessus’ lab was originally forwarded to me by my readers and reported on my site, I then posted it also on PubPeer. Under mounting pressure from her peers, Jessus started to correct some of her papers, and eventually her employer CNRS, Europe-largest network of research institutes, decided to switch to attack mode of defence. Sorbonne University in Paris, where Jessus holds a professorship, delivered an outrageous whitewashing report about Jessus’s data manipulations, which were either declared to be minor errors, or denied outright, or even declared to be actually good scientific practice, all despite absence of original data. The authors of that opus hid their identity, but one was eventually exposed by my sources: Francis-André Wollman, an elite CNRS and Sorbonne biologist. I was personally accused of libellous slander in a CNRS press release, which also announced to persecute those raising misconduct allegations, while urging French researchers to “collective vigilance” against such traitors. CNRS new president Antoine Petit, whose predecessor Anne Peyroche was deposed after my reporting about data manipulations in her own papers, then announced to smoke out anonymous PubPeer commenters using IT tools of deep surveillance.
Yesterday, an Open Letter has been published on the website Sauvons l’Université ! , authored by around 10 unnamed French biologists. It plucks apart the Sorbonne report of Jessus papers and declares:
“We, geneticists, biochemists, cell and molecular biologists, have read this report, have analysed, criticised it point by point and can conclude that all the arguments of the Inquiry Commission are against all common rules of good scientific practice and of the interpretation of results, which form the basis of research integrity. Worse, and that is the reason that led us to react (see our open letter), it encourages image manipulations which are nevertheless recognized as fabrications and falsifications of results”.
The full version of the Open Letter, which was finalised already on 18 April 2018, was linked under the Sauvons l’Université ! article from yesterday. It meticulously explains to biology outsiders like President Petit what a western blot is, what the correct presentation of western blot results is, followed by what is actually data manipulation and what is fraud. It goes on to analyse one by one the issues in 11 individual Jessus papers as posted on PubPeer, finds virtually all of them to be evidence of research fraud, while denouncing the whitewashing findings of the commission, which the authors declare to be uttelry incompetent in biology. That will sure hurt the pride of cell biology professor and EMBO member Wollman, who, being a pompous fluffed-up chicken, is still afraid to admit his (by now well established) key role in having penned such a disastrous report.
Also the authors of the counter-report prefer to remain anonymous, which is understandable given the threats and calls for denunciation of traitors, publicly issued towards them by Petit and CNRS. This is the full title of the Open Letter:
Lettre ouverte d’un groupe d’experts anonyme concernant le rapport de la commission anonyme ayant enquêté sur les publications dont Mme Jessus est co-auteur
But the title of the pdf document is simply: rapportfraude.pdf. Its authors name the disastrous Jessus investigation by CNRS and Sorbonne University as the reason for having written the Open Letter:
“The conclusions of this report, which at first glance seem to absolve Ms. Jessus and her co-authors of all scientific fraud, are diametrically opposed, as will be seen below, with the findings of the detailed article-by-article analysis of the same report.
Indeed, the inquiry commission finds and admits almost all the manipulations images highlighted on PubPeer. On the other hand, it does not admit that these image manipulations can be fraudulent and justifies them one by one, using
arguments that could mislead by appearing fair and objective to non-specialists.
However, for biology specialists using the same techniques, the entire Inquiry Committee is not only confused, but also scientifically aberrant. The levels of confusion and aberration are such that they reveal the obvious incompetence of this commission of “experts” and discredit its from giving a serious opinion.
The incompetence of the Commission has, in the context of this investigation, consequences even more serious than its own discredit or that of the institutions for which it has done its work. The Commission recommends continuing to do this type of image manipulation. Now this type of manipulation of images, as we explain below, can not be accepted in any case, during a meeting in a laboratory, or as part of a presentation at a congress or as a report in a peer-reviewed journal, such as those the community of biology researchers publishes in.
In this open letter, after a preamble for non-specialists in Biology, our group of experts, consisting of geneticists, biochemists, molecular and cellular biologists, engaged in critical and scientific reasoning on each of the 21 figures resulting in 20 conclusions diametrically opposed to those of the commission of inquiry”.
In the press release at Sauvons l’Université ! the anonymous biologists highlight the main problem, which is not Jessus herself, but the rotten structures of research integrity at the very top of French science. They accuse CNRS and Sorbonne University of “concealment of fraud”, despite having all the tools in place to ensure adherence to good scientific practice. They describe CNRS’ own ethics committee COMETS as corrupted:
“Anne Peyroche was indeed a member of COMETS, before her appointment on 24 October 2017 to the Presidency of the CNRS and her interim advance replacement by Antoine Petit on 18 January following the alleged fraud revealed by PubPeer. Would she have helped to conceal, as a member of the COMETS and then president of CNRS, the fraud carried out in the lab of Catherine Jessus, in order to downplay the same thing conducted in her own lab? Is fraud so widespread that two people at the head of CNRS are implicated? Or is it fraud a means of access to positions of responsibility at the CNRS?”
As I heard, that cartoon Stalinist Jessus was in the process of organising a purge at CNRS. No executions yet, just sackings and demotions of traitors. The danger and fear are very real, and even those in highest positions at CNRS cannot feel safe from Jessus’ revenge. So far, she enjoys the full support from the CNRS president.
Petit is now well advised to stop threatening the scientific community and drop his misguided support of dishonest researchers at CNRS. Otherwise he will be deposed just like Peyroche was. The revolution is on its way.
Below I have now Google-translated and illustrated the French text of the conclusions in this counter-report.
With the exception of comments and conclusions on the two figures of paper 9, all the arguments and conclusions of the commission leave us all very perplexed as to the rigour and professionalism with which it conducted its investigation.
All allegations of PubPeer on any Western blot image manipulations or autoradiographies, the work of such an inquiry commission must be to compare the original Western blots and autoradiographies with the incriminated published figures.
Therefore work must begin with the search for the original documents in the archives of the laboratory. However, it seems that the Commission has merely, as it states in the conclusions of its report ( “The commission asked Mrs. Jessus, to provide original documents used in the assembly of the incriminated figures”) put a request to Mrs. Jessus, who then is said to have provided less than two-thirds, since the commission says at least 7 original documents (of 21) have not been found. At the same time, the national code of ethics of the research practice clearly states that “All raw data (belonging to the institution) and analysis of results must be retained to allow verification”. It is strange that an commission of inquiry into the scientific integrity would make no comment on the absence of such a large number of original documents and does not search for it itself, in the various lab books in laboratories of the authors of incriminated papers.
In any event, the investigating committee takes note of and agrees with almost all the image manipulations reported by PubPeer, namely a non-homogeneous attenuation of background noise (article 8 figure 4D) and 18 image re-uses (of cut-and-paste type) in articles 1 to 8 and 10 to 11.
The Committee recognises that non-uniform attenuation of background noise is “not consistent with good publication practices”, however it does not admit that 18 instances of images reuse can be intentionally fraudulent. As these 18 images falsifications were characterized by the commission, it used a confused and scientifically absurd argumentation, trying to justify the majority of them, after they have been classified into three distinct categories:
– classified as first category, reuse (or duplication) of control samples images in several figures (sections 2, 3, 4, 5 and 7). The commission says such reuses are “legitimate” and claims that “there erroneous reuse was not intended” or that “the control conditions of each experiment were reproduced.” It explains that it is “justified to reuse controls in several figures when these are from the same samples” and goes on to even mention three more figures with images of reused control samples, than reported on PubPeer (in section 5).
However, the Committee obviously confuses reuse of control samples and copy / paste of images of control samples , which is not at all the same thing! In the first case, reuse is scientifically correct, it is to rerun the control samples while performing an additional Western blot with other samples from the same experiment. In the latter case (which is the case of incriminated articles), it is scientifically not correct. Indeed, the control samples were not subjected to a second Western blot, only their images were recovered and pasted with the images of other samples in the experiment, making the resulting figures uninterpretable! This confusion of the Commission of Inquiry, which made it wrongly and re-iteratively conclude “that there is neither scientific misconduct or error, there is no doubt of the scientific message of the papers”, highlights its total incompetence.
– classified in the second category of image manipulations are not related control samples and of pasting (or inserting) parts removed from the same blot membrane (papers 8 [Figure S4] and 11) or different blot membranes (papers 1, 6 [Figure 5C] and 11).
We have explained many times (see Figure 5 of the preamble and our comments on papers 1, 6 and 11) that the latter type of assembly (splicing of gels or different membranes) is scientifically not rigorous since the samples which images are artificially moved have not been subject to the same western blotting or autoradiography, they are not comparable, which makes them uninterpretable figures. The commission should have recognised such handling as intentionally fraudulent, instead of justifying it.
On the other hand the assembly of distant parts of the same gel (or a single membrane) is scientifically acceptable (see preamble page 9). However, when one carefully analyses the organization and the noise pattern of the S4 section 8 and S1B and S2B of paper 11, those are not distant parts of a same membrane being assembled (n◦3 case of assembly in the preamble), but assembly of pieces of different membranes (n◦6 case of assembly in the preamble). It is very clearly made to give the illusion that it is the same membrane, which is completely fraudulent.
Again, and despite the observation of certain originals, the commission was not able to distinguish between ‘legitimate’ assemblies and assemblies interpretable scientifically as intentional fraudulent manipulation, leading to believe that samples of the same figure were all subjected to the same western blot or even autoradiography.
– classified into the third category are reuses (or duplication) images of samples which are not controls and which are found twice in a single panel (as part of a Western-blot of Figure – Figures 3 and 5B of the article 3, figures 6 and 7 of the paper) or reuse of control samples images that are found in two different panels (paper 10).
While it is exactly the same type of copy / paste of fraudulent images as in previous cases, the Committee recognises that this time they are errors. However, by fallaciously reasoning that a panel can be made from different assembly parts (as is it was a kind of a puzzle), the commission downplays those errors by passing them as unintentional errors, unfortunate blunders that could have led the authors to use wrong parts when assembling the panel! But here the error is not to have mistaken a piece during assembly, but to perform such assemblies in the first place! In addition, the Committee points out that these errors were corrected. However, to date only 4B of paper 10 is the subject of a corrigendum published by J. Cell Science. One wonders what did the commission try to achieve by performing such a ranking of the 18 image manipulation in three categories? Clearly to try to exonerate 13 of them and minimize the fraudulent intent of the other 5.
This attempt is vain, as we have shown, article by article, figure after figure, the 18 images reuse are all forgeries, to make believe that the samples of the same panel underwent the same Western blot for some, or to hide an original result for others.
Since the Commission refers to this repeatedly in its analysis, it should be noted that the vast majority of peer reviewed journals in Biology publish for several years in the instructions for authors a paragraph on which image manipulation is acceptable or which is not for publication (for review see [Rossner and Yamada, 2004] and Annex I). This paragraph was non-existent in the 1990s, appeared in the last fifteen years and has become increasingly detailed in order to try to reduce cases of fraud identified by publishers.
However the advice recommended in such paragraph is not to become a law, with no retroactive effect, to determine what is good or bad scientific practice, since good or bad scientific practices are independent of the year of publication of this paragraph in journals. Good or bad practices are in fact determined by the limits of interpretation of each technique and each experiment as explained in the preamble. Realising, analysing and rigorously presenting every experiment within these limits is a good practice (i.e., for the technique of Western blotting, all samples to be compared must have been subjected from the beginning to the end to the same Western blot), violating those limits demonstrates a lack of rigour and is a poor scientific practice.
Contrary to attempts by the commission to insinuate throughout its analysis, in order to exonerate the perpetrators, is that even if good scientific practice is not specifically formulated in the recommendations of the journals to authors, this does not allow non-compliance with this good practice. Failure of good practice is bad practice no matter when it happens, in 1997 as in 2017!
The argument put forward by the Commission in papers 1, 2, 3, 4, 5, 6 and 7 is particularly disconcerting. For these papers, the commission concludes its analysis by stating that nothing challenges in the scientific messages of the paper. On this occasion, we would like to tell the commission which does not seem to be familiar with the experimental sciences, that a scientific experiment is designed to answer a question and the answer to this question is given by the raw data (or original) of experiment and this alone. For papers 1 to 7, the original documents were precisely not found, or not all have been found. On what basis does the commission declare that the scientific message of the paper is not altered if there is nothing to consult?
On the contrary, the accumulation of fraudulent practices in a laboratory over a period of 20 years, combined with a significant loss of original documents, should seriously alarm the inquiry as to the scientific message given by the articles.
In conclusion, all the comments of the inquiry commission show that it absolutely does not master the techniques used in the papers, that it neglects the fundamental importance of internal controls of each experiment, that it confuses legitimate panels assembly with illegitimate assembly of subpanels, and finally, it confuses unintentional errors with intentional manipulation meant to deliberately falsify an original result.
All arguments of this commission of inquiry, which are more like an exercise in rhetoric or dishonest advocacy, have nothing to do with scientific arguments. They are irresponsible because they encourage image manipulations. They can in no way be considered either by the institutions or by the journals Developmental Biology, Journal of Cell Science, Development, Molecular Biology of The Cell and Cell Cycle.
Update 18.05.2018. I now got in touch with the anonymous authors of the counter-report. This is their message:
“We are about ten researchers from different universities or institutes. […]
Moreover, this crisis is by far the most serious situation we have ever encountered. We absolutely need the protection of anonymity to continue to communicate freely without pressure and without threats to our careers. This will also ensure that the institutions cannot follow their usual strategy of sueing for defamation and attempting to discredit the whistleblowers. They will be obliged to confront the real problem.
For this reason we have provided only information that can be verified factually. We have not included any unverifiable statements or opinions”.
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. Your generous patronage of my journalism, however small it appears to you, will greatly help me with my legal costs.