The authors of a 2012 paper in Cell have retracted it after discovering “serious issues with several figures.”
Here’s the notice for “A PLETHORA-Auxin Transcription Module Controls Cell Division Plane Rotation through MAP65 and CLASP:”
Our paper reported that Arabidopsis PLETHORA transcription factors regulate cell division planes by transcriptional activation of MAP65, which interacts with the CLASP protein to guide microtubule orientation. We recently identified mistakes affecting Figures 4N, S4B, and S6E in which original data were processed inappropriately such that the panels do not accurately report the original data. At least in one case, the original data did not support the figure’s conclusion. We believe that the most responsible course of action is to retract the paper. We sincerely apologize to the scientific community for any inconvenience that this might cause.
The paper has been cited 27 times, according to Thomson Scientific’s Web of Knowledge.
The authors of the paper span labs in countries from Canada to Japan. One of the corresponding authors, Ben Scheres, of Wageningen University in the Netherlands, tells Retraction Watch he was alerted by a colleague to “suspicious anomalies” in the paper on April 24 of this year:
I immediately contacted my co-authors, Cell, and Utrecht and Wageningen Universities about the concerns regarding the paper. We performed our own investigation and identified serious issues with several figures.
The co-authors and I have requested an official retraction of the Cell paper, which has now been posted. Further independent investigations about the cause of these issues are still ongoing, and I cannot yet inform you about their outcome. Once the investigations are completed, I will however be in a position to provide you with details of the case and try to answer questions that you may have.
Please see an update on this post.
A SNAFU or worse? No comments on PubPeer apart from a statement that the paper is now retracted, so a sharp eyed reader spotted the problems and send their concerns to the authors. At least the authors seem to be “doing the right thing” after identifying the problems.
It will be worse. In the field of plant sciences extremely serious concerns of image manipulation in other papers of the same team exist. Surely, this is the first of many in top journals. I bet that the next retracted paper will be this: http://www.ncbi.nlm.nih.gov/pubmed/23515161.
I think that the recent editorial in plant cell/plant phys about manipulation of image data (http://www.plantcell.org/content/25/9/3147.full) was not by coincidence.
The retraction statement has been changed last minute without any of the coauthors consent. Specifically, Cell and the first author switched “manipulations” to “mistakes”, which is completely inaccurate and misleading. The manipulations were performed by the first, second, and sixth authors, without any knowledge by any other authors, so the statement is totally misleading. Cell will now be investigated for a deliberate cover-up, and the media will be contacted.
Not sure about a sharp-eyed reader. All the retracted figures are RT-PCR, and look perfectly fine.
Something odd is happening given that the first author and co-corresponding author Pankaj Dhonukshe no longer has a web page. He got the most prestigious award possible in Europe: the ERC grant. 1.5 million euros. But try following the link to his lab home page from the ERC website.
http://erc.europa.eu/erc-stories/decoding-origin-plants%E2%80%99-formation
They no longer list him as faculty. Very odd. He also has numerous other high profile papers.
In fact, the Nature paper from Scheres and Dhonukshe in 2013 mentions the same genes, CLASP and MAP65, as the Cell paper that was retracted. I am not sure if the same alleles are used in the Nature paper as in 2012 Cell paper because it is behind a paywall. But the Cell retraction was based on the RT-PCR to show alleles of these genes were real knockdowns.
RT-PCR in Fig. 4N don’t look perfectly fine to me.
Firstly there are three results in each lane: Actin, MAP65-1, MAP65-2. This is unusual as each PCR amplification experiment would normally be done independently.
Secondly, it is unlikely that the left-hand lane Actin and MSP65-2 bands could be from the same gel lane as the band shapes are so different.
Thirdly, in the 3rd lane, the MAP65-1 and MAP65-2 bands are superimposable.
So sharp-eyed reader is perfectly possible. Though in this case we are being told by one of the authors (comment below) that the sharp-eyed readers were also co-authors.
Cell seem to have taken the supplement off line so I can’t comment on the other listed problem figures (S4B, S6E).
Look more closely at those figures (i.e. throw them into imagej, adjust contrast, then prepare to laugh), they are very sneaky and have fooled many.
Here are the supplementals:
https://www.dropbox.com/s/mqow9rkabowqkbf/Dhonukshe%20nature%20CLASP%202013-SUPPs.pdf
Thanks!
It is surprising that S10k is not listed above, the splicing is so rough that the edges of the PIN2 gel are ragged where the crops don’t line up. What is the purpose of the thin black strip between lanes 1 and 2?
This actually is the nature paper (wonder when that will be retracted) not the cell paper.
Oops, wrong one. Here is the cell paper with supplementals:
https://www.dropbox.com/s/qcndedxt239xrxr/dhonukshe%20cell%20clasp%20map65%2Bsupps.pdf
A bit of the analysis showing the manipulations:
https://www.dropbox.com/s/vnsdb2zwaar7lq5/Dhonukshe%20cell%20figures-2.pdf
Note: The author of the comment has deleted the PDF at the request of Utrecht.
Kudos, that is impressive detective work. I didn’t notice anything wrong at all.
In fact two of the co-authors also spotted the anomalies and reported them to Cell back in April after we saw clear manipulation of data in the 2013 Nature paper on which Dhonukshe is corresponding author and decided to see if similar manipulations occurred in the Cell paper. Suffice it to say, we don’t feel that these are simple mistakes. Prepare to be both shocked and disgusted once the full story comes out.
I am shocked that anyone would fake RT-PCR for knockdowns using photoshop. The way to do it is just use a small number of cycles.
These guys were amateurs
The leading researchers of this paper knew about Dhonukshe’s reputation. The 2003 PPB paper in Plant Cell has calculation “errors”, which were already pointed out to the authors before publication.
2003! How many of PD’s papers are suspect?
As a co-author in the Cell paper, I am wondering why you did not see the clear manipulation of data immediately after you received the first draft. That would be a great contribution to the society. However, it looks like that you decided to wait. Ironcially, you are stilling keeping the Cell paper in your publication list on your webpage. Probably we should read through your publication as well since we do feel that there are ‘simple mistakes’ in your papers as well.
[Comment deleted at user’s request.]
[Comment deleted at user’s request.]
jw: you are right about the first author. The corresponding author mentioned that the investigation is going on.
Who want to bet that the microscope images are also picked to fit the story?
I wonder what the sixteen other authors in the paper (other than Dhonukshe and Scheres) have to say about this. Didn’t any of them notice these anomalies? Surely their permissions were taken before allowing their names to be put on (a potentially career terminating) paper like this one?
In any case, it is unfortunate that Dhonuske’s scientific career is down the toilet now. Talk about falling from grace. An ERC grant one day, and total shame the next.