Research integrity Research Reproducibility

How Emilie Marcus and Cell covered up misconduct at Weizmann

by Leonid Schneider
Comments 13

I previously reported about numerous cases of suspected (or even blatantly obvious) data manipulation at the Weizmann Institute of Science in Israel (here, here and here).  Initially I wrongly assumed that the institute does not investigate misconduct evidence on principle. It turned out they do, but these investigations “are not public” as Michal Neeman, Vice President of the Weizmann Institute, told me in an email.

Below I present the documents from one such investigation from 2012, regarding two papers by the Weizmann cell biologist, apoptosis researcher and keen dancer Atan Gross.  The original report was filed in July 2012 by a peer of Gross, David Vaux, deputy director of The Walter and Eliza Hall Institute of Medical Research in Melbourne, Australia. The Weizmann investigation ended just one month later with data manipulations confirmed, original raw data absent, but with the conclusion that it all was only an “unfortunate decision about presentation of the data” with “no evidence for falsification with an intent to deceive”. The case was closed with the request of two corrections, one to Journal of Biological Chemistry (JBC), and another one to Cell, with the manipulated data replaced by newly generated Ersatz. While the former journal duly issued a correction, Cell‘s Editor-in-Chief and Cell Press CEO Emilie Marcus declared that she will do nothing at all.

stress in stem cells text

This is exactly same attitude to research integrity Marcus showed some years later, in the case of Olivier Voinnet and Maria Pia Cosma (here). PubPeer is full of examples of unresolved issues with Cell Press papers, even a paper by most notorious zombie scientist Susana Gonzalez remains safe. Condoning research misconduct and fooling readers into believing in occasionally non-existent data integrity of Cell papers became unofficial policy under Marcus, while a Cell paper remains a safe ticket to tenure or a huge grant for its authors, as it was probably for Gross also.  It is not clear if this policy will change with the new Cell leadership: Marcus now left Elsevier and became a university bureaucrat with the David Geffen School of Medicine at UCLA.

 

sd2bbni.png
Duplicated (and unspliced) loading controls aside, the gel lanes in both BID panels were stealthily spliced. Weizmann decreed that this practice was “still acceptable at the time the two papers were submitted for publication”. Meaning, Weizmann regards loading controls as irrelevant poppycock which noone needs anyway.

I highlighted some of the issues in Gross papers in my earlier article. The two papers Weizmann investigated in 2012 were:

Rachel Sarig , Yehudit Zaltsman , Richard C. Marcellus , Richard Flavell , Tak W. Mak , Atan Gross

BID-D59A is a potent inducer of apoptosis in primary embryonic fibroblasts

Journal of Biological Chemistry (2003) doi: 10.1074/jbc.M210296200
Iris Kamer , Rachel Sarig , Yehudit Zaltsman , Hagit Niv , Galia Oberkovitz , Limor Regev , Gal Haimovich, Yaniv Lerenthal , Richard C. Marcellus , Atan Gros

Proapoptotic BID is an ATM effector in the DNA-damage response

Cell (2005) doi: 10.1016/j.cell.2005.06.014
xdsjbjz
Duplicated background in Kamer et al Cell 2005, source PubPeer. According to Weizmann without any “intent to deceive”
The evidence of data manipulations was meanwhile posted on PubPeer (here and here), and included in many cases duplications of bands and background patches as  well as uncounted instances of inappropriate gel splicing. Weizmann decided those latter were perfectly acceptable practice at that time, but irregular splicing (e.g., where loading controls are not spliced and hence derived from a different gel) was in fact never acceptable and counted always as misconduct.
krii69m
Duplicated gel lanes in Kamer et al Cell 2005, source PubPeer. Verified as scientifically reproducible result by Gross himself, which Weizmann described as “independently”

“An unfortunate decision about presentation of the data without an intent to deceive”

In fact, the Cell paper contains many more hidden splicing events then discussed in the Weizmann report from August 2012, which I present here. Weizmann also issued a summary to the editors of the two journals, signed by the Vice President Haim Garty, who died two years later. This was the letter from August 27th 2012, the original document is here:

“In accordance with the Weizmann Institute practices regarding allegations of deviations from Responsible Conduct in Research (RCR), I have nominated an inquiry committee whose mission was to determine whether there is evidence of a possible deviation from RCR to warrant an investigation (i.e. a formal examination and evaluation of all relevant facts to determine the responsible person and the seriousness of the misconduct). The committee included two very senior members of our scientific community who are of impeccable integrity. Both are also well familiar with the relevant research area and the findings subject to the allegation. As customary, I have blanked out their names in the attached report. These will however be disclosed to you in confidence, should you find it essential in your own evaluation process.
Prof. Gross was asked to bring all relevant lab notebooks to the Dean’s office and in the presence of witnesses identified the original data subject to the allegation, as well as additional experiments done during the same period which may (or not) support the authenticity of the scientific findings subject to the allegation. The committee examined all relevant data and interviewed Prof. Gross at length. Their findings, which include analysis of every image suggested to undergo illegitimate editing, are summarized in the attached report (The committee’s report). The committee’s key conclusion is that data has indeed been extensively edited. However, these can be classified as an “unfortunate decision about presentation of the data” and there is no evidence for falsification with an intent to deceive.
Based on the committee’s recommendation the Weizmann Institute will not carry further inquiry into this matter. Obviously, you may seek taking additional corrective steps. The committee suggested that a corrective statement drawing readers attention that certain figures have been excessively edited will be published. In addition, corrected figures plus the original images will be uploaded as supplementary material. The text suggested for such statement is:
“Figures X, Y, Z were excessively edited. Corrections to these figures (including the raw data used to generate these figures or additional data confirming the conclusions appear as Supplementary Material“.
The suggested supplements which also demonstrates the editing that took place and the authenticity of the research findings are appended as well (Sarig et al JBC 2003 Supplementary – corrections; Kamer et al Cell 2005 Supplementary – corrections)”.
kuqtefj
From Kamer et al Cell 2005, source: PubPeer. Arrows show splicing (though right and left parts of first gel image derive from two different gels), colour boxes show duplicated background.
These are the two suggested corrections (here and here) which Weizmann sent to the journals JBC and Cell. In most cases, it is clearly stated that the original data was not available for papers which were just 9 respectably 7 years old. Reproducibility was ascertained with Gross faithfully repeating (in a matter of days!) year-old experiments he previously saw the need to Photoshop. Weizmann was satisfied with that:
“Prof. Gross was able to provide films from additional experiments, not included in the published papers, which independently confirm the scientific conclusions of the edited figure”.
Not a single one of his other papers was examined, in fact the unnamed Weizmann experts expressly warned against this:
“Based on these findings we recommend that no further investigative steps be taken by the Weizmann Institute”.
The analysis of these two and several other Gross papers was then done by other readers later on, and posted on PubPeer (see here, as well as here and here). It is not clear if Weizmann did anything about that.
mfalrvm
From Sarig et al, JBC 2003, source: PubPeer. A band proved duplicated, original gel didn’t exist (“The exact autorad was not found”, as Weizmann conceded).
JBC published a correction in October 2012:
“Figs. 1A, and 7, A and B, were excessively edited. Corrections to these figures, including replacement of Fig 1A with an exactly identical experiment and addition of white lines to mark joint lanes in Fig 7, A and B, appear below”.
In fact, why this Figure 1A had to be corrected is unclear. After all, the Weizmann experts decreed in regard to what clearly looks like duplicated bands or bits of background:
“There was one alleged duplication of a lane. Detailed examination of continuity of patterns in the image background suggests that the two lanes in question, albeit very similar, may not be identical. However due to lack of the original film, this issue could not be fully resolved. We note that data from an independent similar experiment confirms the scientific finding.

Background editing was found in a number of figures, and was listed as ‘same’
in the allegations. For a number of these, either the original data or a similar
experiment was found. Such ‘cleaning’ of background is clearly unacceptable
practice. Nevertheless, as best as we can tell, in none of those instances was the conclusion affected by the “cleaning”, and in all cases additional experiments were provided that support the same conclusion”.

Only admit what it is safe to admit, and deny everything else. Exactly this is why misconduct investigations, and the names of the investigators must be made public, otherwise it doesn’t matter in the end if the investigations ever took place or not. But Weizmann strategy is different. Good that JBC issued that correction.
Cell however did nothing at all, and one can rest assured Weizmann did not protest too much. This was an email Emilie Marcus sent in June 2013, to explain why she decided to cover up the evident misconduct:
“We have investigated the matter extensively with both Dr. Gross and the Weizmann investigation committee.
Because of the length of time that has elapsed since the original publication, it was not possible to identify the original data in many cases and reconstruct the published figures. Dr. Gross did provide new data to corroborate the original findings but we did not feel that this was relevant or valid to resolving concerns about the original published figures. We also however did not find sufficient evidence to compel a Retraction.  For this reason we have considered all options and concluded that there is no reasonable and appropriate action for the journal to take at this time.
I appreciate that this outcome is not likely to be satisfying to you, but the specifics of this situation with the timing, the nature of the figure issues, and the findings of the investigation committee leave us with limited options.
Thank you for bringing these concerns to our attention. We remain strongly committed to ensuring the accuracy and ethical standards of what we publish.
Best wishes,
Emilie”

 

 

Donate!

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!

€5.00