Delays, arguing over upcoming Cell retraction leave first author “devastated”

After being “blindsided” a few months ago when she was told one of her 2005 papers was going to be retracted, a researcher scrambled to get information about why. And when she didn’t like the answers, she took to PubPeer.

Eight days ago, Shalon (Babbitt) Ledbetter, the first author of the 2005 paper published in Cell, posted a comment on the site announcing the paper was going to be retracted after the last author’s institution, Saint Louis University (SLU), determined that some figures had been manipulated by the last author, Dorota Skowrya. A letter dated September 2, 2015 sent by SLU to Cell describes the results of the investigation — namely, that the manipulations were “cosmetic,” and had no effect on the data or the conclusions. More than two years later, Ledbetter learned the journal was planning to retract the paper, and an initial draft of the notice wouldn’t identify who was responsible; she has since been pulled into a confusing web of blame-shifting and conflicting information that has been, in her words, “heartbreaking.”

Ledbetter wrote on PubPeer:

Cell claims that this paper should be retracted to “correct the scientific record,” yet they are willing to jeopardize the reputations of innocent parties and disregard documented facts by refusing to set the record straight. Because of this baffling inaction by Cell, I am forced to defend my integrity against any potential backlash from the upcoming release of this retraction statement.

“Cleared of research misconduct”?

Cell declined to answer our questions about the paper, and Skowrya did not respond to requests for comment. In response to our questions, SLU provided us with a statement from Matthew Christian, associate vice president for research:

Saint Louis University has been informed by the journal that the editors intend to retract the paper. In 2015, SLU completed an investigation of allegations of research misconduct against a SLU author of the paper, and findings were shared with the editor of Cell and study co-authors at that time. The researcher was cleared of research misconduct by SLU and DHHS. As reported to the journal editor and to co-authors in 2015, the investigation found alterations made to images in the paper were cosmetic in nature and did not affect the interpretation of the data or conclusions drawn from it.

An SLU letter dated September 2, 2015 to Cell editor Emilie Marcus paints a different picture of the university’s conclusions — namely, that Skowyra’s manipulations were “consistent with research misconduct:”

While the committee found that Dr. Skowyra did engage in image manipulation as alleged, a practice that they deemed consistent with research misconduct, they also concluded that the changes made to the images were “cosmetic” in nature and neither affected the interpretation of the data nor the conclusions drawn from the data.

Why Cell decided to retract the paper more than two years after this letter remains unclear. “ATP Hydrolysis-Dependent Disassembly of the 26S Proteasome Is Part of the Catalytic Cycle” has been cited 96 times, according to Clarivate Analytics’ Web of Science.

Going public

When we asked Ledbetter — now at Pfizer — why she decided to take to PubPeer to express her concerns, she told us she had “just happened upon” the site. “I had no idea such a forum even existed until a couple of weeks ago.”

She continued:

I posted my concerns because Cell is refusing to include accountability in the retraction notice when they have official documentation from SLU saying otherwise, and I cannot stand by and let my reputation and the reputations of the other innocent coauthors be treated so recklessly.  When I spoke on the phone to the Cell editor a few weeks ago regarding the language they plan to use in the retraction notice, she explained that the position of Cell is that the readership will understand that any misdeeds reported in the notice are inherently the responsibility of the corresponding author.  I disagree with this position, therefore I sought another avenue to voice this dissension, and found PubPeer.

Ledbetter forwarded us a series of emails, including one from Cell to the authors on February 15, 2018, in which the journal says the errors cannot be fixed with a correction:

In multiple previous communications, we have explained to you that the nature of the data manipulation issues reaches beyond the acceptable standards held by the community.  Our expectation is that the data that are provided, peer-reviewed, and published in Cell Press journals accurately represent the data obtained in the lab at the time and need to conform to community standards and expectations for rigor, accuracy and credibility. In this case, the problems with data falsification, both stamping and lane duplication, have compromised accurate communication of the original data to the reader. It is simply not possible for Cell to acknowledge this degree of manipulation as an acceptable standard, and we continue to see a retraction as the only way to correct this for the community.

We had hoped to convince you that the best course of action both for the scientific community and for your coauthors would be to retract the paper in your own words.  As you are not prepared to take this step, we must move forward with the editorial retraction.

In the Feb. 15 email, Cell included a draft text of the retraction notice, which notes:

An investigation [at SLU] determined that data in three figures had been inappropriately manipulated.  Figures 1C, 2B and 4D contain evidence of stamping, placing a cut-out section from one part of a processed image over another part of the same image. In Figure 1C, because the stamping covers bands for two regulatory complex components (Rpn12 and Rpn10) in a control condition, support is enhanced for the conclusion that complex dissociation occurs under conditions that support ATP hydrolysis.  The stamping in the other panels did not obscure data. In addition to the three panels where stamping occurred, data were also falsified in Figure 2A where one of the lanes in the Western blots was duplicated… As a consequence of these manipulations, the published figures are not accurate representations of the original data. We have therefore concluded that the most appropriate course of action is to retract the paper.  The authors do not agree to the Retraction.

Ledbetter replied, asking the journal to modify the text to make it clear Skowyra was responsible — noting on Feb 18:

As it stands now, the language is unacceptable …

In response, the editor suggested they speak by phone.

“Blindsided” by news of the retraction

Ledbetter first learned of the retraction via an email from the same Cell editor dated Friday, Nov. 3, 2017; on Monday, Nov. 6, she responded:

This revelation has been heartbreaking and I have been completely blindsided by this decision, as I have not been informed of the events that have led to this retraction notice.  I am currently in the process of contacting SLU to get more information. As the first author, I am absolutely shocked that things have progressed to this level without my knowledge.

I was informed in August 2015 by Dr. Skowyra that anonymous allegations of misconduct were made against certain figures in my paper, but I was also informed in June 2016 that SLU concluded that no misconduct had taken place. So to receive a notice of retraction nearly a year and a half after supposed vindication is baffling and distressing … Please understand that I do not take this outcome lightly as I believe my research has always been above board.  I am truly devastated.

On Nov. 15, Ledbetter contacted Skowyra, asking if she is willing to modify the language in the notice to make it clear that Skowyra is responsible for the manipulations (Ledbetter’s emphasis):

… as the first author, I am particularly vulnerable to such speculations without a clarification of responsibility for the described manipulations.  I can’t speak for the other coauthors, but I am at a critical point in my scientific career. I am planning to transition out of my postdoc soon and, as I’m sure you understand, a retraction may have a significant impact on future funding and job prospects.  I understand that you are against this retraction. But, like I said above, the retraction is inevitable. In my view, the just path is to attempt to mitigate damage of this retraction to the reputations of the faultless coauthors, including myself. Considering this, are you willing to take full responsibility for the image manipulations described in the statement? For example, including a sentence like “Dr. Skowyra has acknowledged engaging in these image manipulations.” This is a no-win situation for all of us, but it hardly seems fair to penalize those of us who had no knowledge of any impropriety, cosmetic or not.  To me, this represents “research integrity.”

On Nov. 16, Skowyra responded, noting that she too is “distressed” by the situation (her emphasis):

… as you noted, many unnecessarily detailed and conflicting statements have been released by the University, including the statements that Cell press now proposes to misuse. This list also includes the statement that it is me who is responsible for the image manipulations, which is incorrect – the original investigation report clearly states that the manipulations were done by a person or persons unknown. I was the respondent because of my role as the senior author, not as the person who has done the image manipulations. I am responsible for letting the figures be published as they were, but so are all coauthors – each of you have approved the submission. And I took the responsibility of the respondent very seriously, which has been acknowledged in the investigation report and which you know, because I have sent several updates about the investigation.

“Cell decided to act on its own”

Ledbetter also reached out to SLU in early November, seeking more information about the case. On Nov. 7, she received an email from the vice president for research:

Cell contacted Dr. Skowyra earlier this year, asking her to voluntarily retract the articles due to the manipulation of Western blot images. Cell’s stance is that the image manipulation required a retraction regardless of the underlying impact on the science itself. When Dr. Skowyra refused to voluntarily retract the article, Cell decided to act on its own asking SLU to confirm that the words used matched the findings [of SLU’s investigation]. While it is true that Cell is using the language from the letter sent to them in this retraction, they did not accept Saint Louis University’s recommendation that they include the full finding that “the alterations did not affect the reported findings, or their interpretation.”

The retractions are unfortunate because we have no evidence that the underlying science was altered, but the manipulation of images, which seems to be the journal’s concern, is not in dispute.

Less than 30 minutes after receiving SLU’s email, Ledbetter forwarded it to the paper’s co-authors, leaving out Skowyra. She noted:

In reading this retraction notice, I would most likely be presumed the “guilty” party, since my name (i.e. Babbitt et al.) is the only one that is explicitly stated.  Although I was a research technician at the time of this publication, having limited control over figure preparation, no one would know the extent of my contributions to the elements of the paper based on this notice.  While I was in Dorota’s lab, I provided her the original westerns on autoradiographic film, not fully assembled, processed figures. And of course, I would not have access to any of this data now.

As you all know, the stigma attached to a retraction has prolonged repercussions, and I am concerned about my name and my reputation as a scientist.