About these ads

Retraction Watch

Tracking retractions as a window into the scientific process

“Not exoneration”: University reverses sanctions on husband-wife team found to have manipulated images

with 7 comments

Maria Bravo, via UNAM

Alejandra Bravo, via UNAM

A complicated case involving two microbiology researchers at the National Autonomous University of Mexico (UNAM) appears to have come to some conclusion.

In November of last year, we reported that

…Alejandra Bravo and Mario Soberon, a wife and husband team who study the Bacillus thuringiensis (BT) bacteria toxins used in GMO crops to fight pests,were found guilty of “manipulaciones inapropiadas y categóricamente reprobables” — which translates roughly, according to Google Translate, as “inappropriate and categorically reprehensible manipulation.”

Bravo — who won a 2010 L’Oreal-UNESCO Award for Women in Science for her work on BT — has resigned as chair of UNAM’s Committee on Bioethics, while Soberón gave up the chairmanship of the department of molecular microbiology, according to the newspaper.

The university found evidence of manipulation in 11 papers, and two of those have been corrected. One such notice:

Recently in a review of our papers we found that Panels A and B of Fig. 5 were edited with Photoshop. The published figures did not made clear that they were merged figures, since some lanes of these figures came from different gels. New Fig. 5 constructed with the original gels are given below. This should be considered as definitive by the reader.

None of the papers has been retracted. Now, as reported by La Jornada and then Science, the university has ended the punishments:

In the latest twist, UNAM’s ombudsman Jorge Carmona has lifted the sanctions, citing irregularities in IBt’s investigation. According to Carmona, the problems included the presence of the complainant on the internal IBt committee that first evaluated the case, a lack of adequate opportunities for Soberón and Bravo to argue their position, and a breach of confidentiality that led to rumors and press reports that Carmona says may have unduly damaged the researchers’ reputations. He also says the committee convened by his office was concerned that the punishment was too severe for image manipulation.

But:

“It’s not exoneration,” Carmona says. Rather, his office considers the pair’s punishment “fulfilled” after 1 year.

Bravo and Soberon initially declined to respond to Science, but posted a comment a few day after Science‘s story ran. In it, they note that two journals accepted their corrections “without further actions,” and that one of those journals appointed Soberon as editor a few months later.

They also write that

b) ORI/NIH revised the case and concluded that there was no misconduct since “the questioned manipulations of images appear to lack significance as they do not alter the research results. Thus DIO has administratively closed this case without further action.” We have continued economical support by the NIH grant in collaboration with University of California Riverside.

We find that first sentence a bit puzzling, as we did a La Jornada story in April that seemed to be based on the same source as the quote Bravo and Soberon use. That story claimed that the Mexican Academy of Sciences had released a statement saying the ORI had cleared Bravo and Soberon, but we haven’t seen any correspondence from the ORI about the case.

That’s not unusual. In our experience — and by law — ORI (DIO is its Division of Investigative Oversight) can’t make any public statements about cases in which it does not find enough evidence to make a finding of misconduct. That means it only offers reports on about a dozen of the hundreds of investigations it oversees per year.

Neither does it necessarily follow that “there was no misconduct” just because the image manipulations “do not alter the research results.” Whether that pushed these allegations down the priority list, or whether the university investigation was full of holes, are other questions, but manipulation of the sort described by those corrections is manipulation. So ORI may have closed the case for any number of reasons, but “concluded that there was no misconduct” just doesn’t sound like something we’ve ever heard the ORI say. Simply put, ORI does not exonerate. If anyone has a copy of the ORI’s findings, we’d welcome the chance to review it.

Bravo and Soberon continue:

c) All committees and Journal responses received indicate that there is no evidence of misconduct.

That seems a bit vague, and in the case of journals not particularly surprising. Journals frequently say it’s not their jobs to investigate fraud allegations, and It’s the rare journal that pushes back when an author or institution says there wasn’t misconduct.

UNAM is now “creating a system for handling misconduct allegations,” according to Science — something we’re guessing the university would have been required to have as part of an assurance agreement with the ORI had they been receiving NIH funds directly.

About these ads

Written by Ivan Oransky

October 28, 2013 at 12:45 pm

7 Responses

Subscribe to comments with RSS.

  1. You are correct — ORI does not find “no misconduct” when ORI administratively closed a file — however, concluding that some act or product was the result on an “honest error” or could not amount to a “significant departure from commonly accepted standards” under the federal definition and application might amount to the same thing, that it does not fall under that definition,

    Alan price

    October 28, 2013 at 3:04 pm

  2. Lack of due process seems to be a common problem in lots of university investigations. If the university now finds that due process was not respected, and as a result they have to withdraw their findings, that’s egg on their face and maybe they’ll have incentive to do it right next time. I can’t see that an unfair investigative process is any less problematic than forging Western blots, at least if finding “truth” is the goal of both science and the investigation.

    StrongDreams

    October 28, 2013 at 11:52 pm

  3. ” I can’t see that an unfair investigative process is any less problematic than forging Western blots, at least if finding “truth” is the goal of both science and the investigation.”

    If the information here is correct then there was no faking of western blots. All they did was assemble a figure without leaving enough white space to make it obvious that the lanes were from different gels. If the description of the experiment was otherwise correct, then there is no fraud.

    Dan Zabetakis

    October 29, 2013 at 10:37 am

  4. @Dan Zabetakis is right given the information we have at hand – the problem is that the authors did not convey the true source of the data, multiple gels, and gave the impression that there was one gel. A slightly grey area, as a reviewer may well have asked the samples to be run again had the multiple origins of the data been made clear, but that is all.
    This highlights the simple fact that one should not assemble data this way, without making it clear to the reader.

    ferniglab

    October 29, 2013 at 6:28 pm

    • I think it’s very important to consider that Dr. Soberon is the son of a former Dean of that university, which creates a bit of conflict of interest. So I frankly do not think that the investigation of abnormal behavior is not neither fair nor correct, and I believe there was someone who pulled some strings to end this ASAP. In addition, they have questioned only 11 articles and only the results of two of them have been deeply analyzed. I think the best will be to a totally alien to the institution perform the same experiments, under the same conditions exhibited and compare the results. Moreover, to feign insanity. If you want to avoid authors from using software like Photoshop, editors should better ask for the original photographs of the investigation. In the short time I’ve been working in research, I have heard from the mouths of great researchers that everything, absolutely everything in research can be easily altered to support the research work.

      Mr Akiles Brin Ko

      October 30, 2013 at 1:47 pm

  5. The UNAM’s ombudsman complains about the composition of committees that evaluated the case one year ago. However, the mediation procedure, which now reverses sanctions, was carried out by a committee of 13 persons with the following composition: 7 affiliated to IBT-UNAM (including Bravo & Soberón), 3 heads of other UNAM Institutes, the Dean for research at UNAM, the spokesman of UNAM Rector, and the ombudsman. This is really an unbiased committee?

    Moreover, the memo [1] released by DDU (Defensa de los Derechos Universitarios) is signed by… one person (the ombudsman). Within the Mexican bureaucracy, that means that agreements reached in conciliation are endorsed by the committee, although nobody wants personally to endorse these agreements. In other words, the UNAM forgives to the UNAM.

    [1] http://www.ddu.unam.mx/DDU/Documentos/ComunicadoSep2013.pdf

    Sylvain Bernès

    October 31, 2013 at 9:56 am

  6. You just have to examine a recent paper from the pair Soberon-Bravo in Environmental Microbiology (2013) 15, 3030 (IP>5). All gels are Western blots, not a single one is a simple SDS-PAGE.

    BillXX

    December 14, 2013 at 3:08 am


We welcome comments. Please read our comments policy at http://retractionwatch.wordpress.com/the-retraction-watch-faq/ and leave your comment below.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 33,824 other followers

%d bloggers like this: