Cell Press won’t retract papers despite one author confessing to fraud

Screen Shot 2016-08-03 at 18.54.35Cell Press journals will not be retracting two papers that were flagged with expressions of concern (EOCs) in April after one author claimed to have manipulated some experiments.

In a strange turn of events, as we previously reported, the study’s corresponding author refuted the claims of the author who confessed to fraud, citing concerns about his “motives and credibility.” Since then, two independent labs repeated the authors’ experiments, and “largely confirm” the central conclusions of a Cell paper, but were inconclusive regarding a paper in Molecular Cell. Regardless, in both cases, the journals have decided to take no further action. 

Both expressions of concern (and their associated editorial notes) will remain online, as part of the “permanent record,” a Cell Press spokesperson told us.

The spokesperson added more about the investigation process:

…we asked the corresponding author to provide all the available original data and examined it for evidence that could resolve the conflicting claims. In the absence of any definitive evidence supporting or refuting the claims, we then compelled and oversaw the corresponding author’s arrangement to replicate the findings in an independent lab. This included approving the two labs as independent and without any conflicts of interest, approving the plan for which specific experiments were to be replicated, and evaluating the resulting data.

When asked why the corresponding author was asked to coordinate the replication process despite his vested interests, the spokesperson said:

We think that authors have a central role to play whenever their data are challenged. For this reason, we gave the corresponding author the opportunity to have the experiments independently replicated.

The spokesperson added:

We, as journal editors, did play a role in the process. For example, we approved of the choice of independent labs, ensuring that the replication efforts were performed by labs with no conflicts of interest. We also helped to keep progress of the replication efforts on track and evaluated the resulting data.

The Cell Press spokesperson declined to reveal the identity of the labs where the replication experiments were carried out, citing confidentiality.

The corresponding author on both papers is Xin-Hua Feng, based at Baylor College of Medicine (BCM) in Houston, Texas. The author who confessed to fraud was Yao-Yun Liang; both papers list Liang’s affiliation as BCM, but a spokesperson from the institution told us Liang left BCM in March 2009. 

Adam Kuspa, senior vice president for research at BCM, sent us this statement:

Baylor College of Medicine conducted its own investigation of the allegations of experiment manipulation when they were raised by one of the investigators. The College found no evidence to support those claims. The independent labs have now reached the same conclusion. The College takes seriously any allegation concerning the validity of research and is confident in its review process. We are committed to ensuring the accuracy of our science.

Here’s the first editorial note, issued by Cell on September 8:

We, the editors of Cell, published an Editorial Expression of Concern (http://dx.doi.org/10.1016/j.cell.2016.03.038) earlier this year regarding issues raised about Figures 2F, 2H, and 3G of the above article. Dr. Yao-Yun Liang, who performed the experiments in question, claimed to have manipulated his experiments to achieve pre-determined results. The corresponding author, Dr. Xin-Hua Feng, refuted the claims of falsification. In the intervening time, two independent labs have carried out experiments attempting to reproduce the data in question. This Editorial Note is to inform the community about the outcome of this process. Those two labs have now completed their experiments, and their data largely confirm the central conclusions drawn from the original figures. Although this does not resolve the conflicting claims, based on the information available to us at this time, we will take no further action. We would like to thank the independent labs who invested significant time and effort in ensuring the accuracy of the scientific record.

This 2006 paper, “PPM1A Functions as a Smad Phosphatase to Terminate TGFβ Signaling,” has been cited 239 times, according to Thomson Reuters Web of Science

And here’s the other editorial note, issued by Molecular Cell on September 15:

We, the editors of Molecular Cell, published an Editorial Expression of Concern (http://dx.doi.org/10.1016/j.molcel.2016.03.026) earlier this year regarding issues raised about Figures 5C and 5D of the above article. Dr. Yao-Yun Liang, who performed the experiments in question, claimed to have manipulated his experiments to achieve pre-determined results. The corresponding author, Dr. Xin-Hua Feng, refuted the claims. In the intervening time, two independent labs have attempted to reproduce Dr. Liang’s experiments. Those two labs have now completed their experiments. Despite good faith efforts from both labs, their experiments were inconclusive. We are now closing our investigation because there is insufficient information to warrant further action and because we feel we have taken the matter as far as we can. While we had hoped for a more definitive resolution, we expect that, as with all science, the field will press forward with future papers, adding knowledge that may provide support for (or evidence against) the specific findings in question. This note is to notify the community of the issues and of the conclusion of the process.

The 2002 paper, “Direct Interaction of c-Myc with Smad2 and Smad3 to Inhibit TGF-β-Mediated Induction of the CDK Inhibitor p15Ink4B,” has been cited 138 times. 

Feng and Xia Lin (first author of the Cell paper and last author of the Molecular Cell one) — also based at BCM — did not reply to phone calls or emails from Retraction Watch.

We’ve previously reported on another retraction of a 2013 paper by Lin and Feng in Cell Stem Cell that did not include Liang as an author. Xueyan Duan, second author of the Cell study, is also a co-author of the paper that was pulled last year.

We asked the Cell Press spokesperson for the contact details of Liang (who previously contacted the journals, according to the EOC notes), but were told:

We don’t have any information to share.

We haven’t been able to find Liang’s current contact details.

We have, however, found an erratum that Liang, Lin and others issued in 2003 for another Molecular Cell paper, “Opposed Regulation of Corepressor CtBP by SUMOylation and PDZ Binding,” published in the same year. It reads: 

In this article (Molecular Cell 11, 1389–1396, May 2003), the authors inadvertently omitted the reference that attributes the finding of CtBP-mediated repression on the human E-cadherin promoter. The correct reference follows: Grooteclaes, M.L., and Frisch, S.M. (2000). Evidence for a function of CtBP in epithelial gene regulation and anoikis. Oncogene 19, 3823–3828.

We’ll update the post with anything else we learn.

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.

13 thoughts on “Cell Press won’t retract papers despite one author confessing to fraud”

  1. A hypothesis may be right or wrong. Manipulating data to substantiate a hypothesis that turns out to be right warrants a retraction. The conclusions may indeed by right and so similar data could be acquired, but that misses the point entirely.
    So cell press really are nailing their colours to the mast of hype, rather than rigour. A shame, since their journals were once good and reliable.

    1. But if the results can be replicated, what evidence is there of manipulation in the first place? Just because someone claims to have done something bad doesn’t mean they did, just look at all the false murder confessions. People are getting too over zealous with the retraction hammer these days.

    2. But was the data manipulated? Maybe the second author had an axe to grind with the PI and claimed to have manipulated experiments but actually didn’t. How to tell?

      On a grander scale, I believe the primary goal of publishing in science is to move the field forward, the rewards to the individuals who publish are of secondary importance. As long as the data are correct, the primary goal has been achieved.

  2. It is totally legitimate fair to ask whether Cell Press has been candid or fully transparent in its correction of the literature. The question I have is whether RW’s headline for this story, i.e., that there was an “admission of fraud,” goes two steps beyond what appears to be immediately obvious here? FIRST, was the admission specific? The interpretation of supposed “admissions” are sometimes the most ambiguous features of institutional investigations to deal with, especially when English is not the ‘respondent’s’ primary language. (I have no idea whether that is the case here.) But in this context, one of the most poignant events I recall was seeing a Japanese post-doc waiting alone for his ORI interview armed only with a Japanese-English dictionary. In contrast, Native-English speakers come armed with a bounty of words at their ready disposal for describing an affirmative act involving falsification of data, such as “manipulation” (if that was used here), “misrepresentation,” “mistake,” “error,” “massage” (the list is endless). And even with a native speaker, there are often flaws in specificity of the “confession,” especially so if the admission stops fact-finding. (Again, I don’t know whether that is the case here.) The SECOND step ‘too-far’ is that it is not clear that this event involved “fraud,” a term in the headline that purposefully serves to provoke reaction rather than reflection.

    1. Thanks for your comment. In the original EOC, the publisher notes that Liang said he “he manipulated the experiments to achieve predetermined results.”

      1. Manipulating data to achieve “predetermined results” can result in potentially “fraudulent” results (if this involves more than cosmetics), but that is different than “fraud.” The latter refers to a criminal offense that rarely fits FFP (so to speak).

        So why does RW’s headline have the scientist admitting to a “noun” (fraud) rather than an “adjective” (fraudulent results)? Headlining “fraud” may be good tactics for blogospherics but it confuses the issues underlying the decision about correction of the literature. QED

        I’m all in favor of RW ‘calling out’ Cell, but the more useful issue to explore was suggested in the comment by Teixiera da Silva; that being ‘what can/could COPE do to get a member(?) journal to enforce COPE’s standards?’

  3. 1) One author has admitted fraud.
    2) “Those two labs have now completed their experiments. Despite good faith efforts from both labs, their experiments were inconclusive. ”
    3) And Cell Press (non-)handling of the Cosma, Voinett, Stancheva etc cases have to show that the top predator on Beall’s list of predatory publishers should be Cell Press.
    4 )If you call yourself a scientist, publish elsewhere.

  4. Where are the results of the replication studies?

    They have performed the experiments, they have the results, CellPress has a publishing platform. Why not show them?

  5. You put a PI with a history of retractions in charge of coordinating & reviewing a replication of a manipulable experiment. That’s brilliant.
    Is “independent” really that hard to understand? Blind samples and controls, dammit.
    CellPress dishonesty/greed has turned them dumb. Pity.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.