One Friday in January, graduate student Meredyth Forbes was reviewing material for her dissertation with her mentor when she decided to make a confession.
She “burst out with a statement that some of the data was fabricated,” said Edward Burns, research integrity officer at the Albert Einstein College of Medicine, where Forbes worked. It was, Burns told Retraction Watch:
a spontaneous, unsolicited, admission to fabrication.
Forbes left her post at Einstein shortly thereafter. Cooking primary data for her research on cells in zebrafish cost her and principal investigator Florence Marlow one paper — the retraction notice was published yesterday — and earned them an expression of concern in another. A correction is forthcoming on a third paper, according to Burns.
Because the U.S. National Institutes of Health supported the work, the Office of Research Integrity (ORI) was alerted to the fabrication, Burns said:
ORI has reached out to the former graduate student, and they have agreed on a way forward….ORI wanted Ms. Forbes to sign an agreement related to future research, and Ms. Forbes signed the agreement.
Marlow told us:
We have been working hard to repeat the experiments in question. We have been in communication with our Research Integrity Office, all coauthors, and the Editors of Development and Cell Reports, and working with them to find the most appropriate resolution and to ensure that the integrity of the literature is maintained.
Here’s the retraction notice for “Maternal dazap2 Regulates Germ Granules by Counteracting Dynein in Zebrafish Primordial Germ Cells” published in Cell Reports:
The first author, a graduate student, has admitted orally and in writing to fabrication of data that she contributed to this article in which we reported a maternal requirement for the scaffold protein Dazap2 (Mdazap2) in germ granule maintenance by a mechanism that is epistatic to Tdrd7 and that counteracts Dynein activity. Furthermore, we reported that Dazap2 binds to Bucky ball, an essential regulator of oocyte polarity and germ plasm assembly, and colocalizes with the germ plasm in oocytes and in primordial germ cells. Although the protein interactions, expression, and localization results are valid, the first author has admitted orally and in writing to fabrication of all “mutant data” in Figures 1H and 1J, Figure 2, Figure 3, Figure 4, and Figures S2 and S3. Because the data presented in those figures do not reflect the standards of quality that are expected in science, and in order to protect the integrity of science, our laboratories, and institutes, we are retracting the paper. All authors agree with retraction of the paper.
The paper, published last July, has been cited once, according to Thomson Reuters Web of Science.
Another paper, “The polarity factor Bucky ball associates with the centrosome and promotes microtubule rearrangements to establish the oocyte axis in zebrafish,” published in Development in December, was flagged with an expression of concern in January:
The corresponding author of this article, currently posted online as an Advance Article, contacted us regarding potential issues with some of the data contained in this manuscript. The authors are currently investigating the matter, in consultation with their Research Integrity Office.
Development is publishing this Expression of Concern to alert readers to the situation while the investigation takes place, and it is not a statement as to the validity of the data. The article will remain posted on the Advance Articles Page until the matter is resolved.
This course of action follows the advice set out by COPE (Committee on Publication Ethics), of which Development is a member.
The paper has not yet been indexed in Web of Science. Marlow told us that its status may change:
At this point it seems there will be a correction, but we are still completing experiments. Because we are still conducting experiments I do not know when this will be published.
Marlow also said she had asked Development to correct a third paper, “Dachsous1b cadherin regulates actin and microtubule cytoskeleton during early zebrafish embryogenesis,” published last August. The correction has not yet appeared online.
In cases where students fake data, it’s easy to wonder whether the supervisor should have been more involved. The school asked that very question, said Burns, and found that a lack of oversight wasn’t the issue:
The mentor and the student met weekly. All the data was reviewed. The data was cooked…[Forbes] admitted that she hoodwinked her PI by showing her primary data that was corrupted.
Burns added that the situation has taken an emotional toll:
It was very sad, within the department, within the school…[there was] outrage at the act, but coupled with sincere feelings of pity… it’s just terrible.
We have reached out to Forbes on Facebook, and will update this post with anything else we learn.
Update, Thursday April 28th 11:10 am EST:
The correction for “Dachsous1b cadherin regulates actin and microtubule cytoskeleton during early zebrafish embryogenesis” has now appeared online. It explains that the authors repeated the experiment for which Forbes faked data, and that their results support the original conclusion of the paper:
In Development 142, 2704-2718, the data presented in Figure 2F indicated that there was a significant reduction in acetylated microtubules in stage Ia and Ib dachsous1b oocytes compared with wild-type oocytes, but that dachsous1b and wild-type oocytes were comparable by stage II of oogenesis. Based on these data and other oocyte analyses reported in the paper, we concluded that dachsous1boogenesis is largely unaffected. The second author (M.M.F.) admitted that, without the knowledge of the other authors, she manipulated the stage Ia and stage Ib mutant data shown in the original Figure 2F. Therefore, we repeated this experiment and examined five wild-type and five dachsous mutant ovaries (>20 oocytes per genotype per stage), and found no significant differences in acetylated microtubules between wild-type and dachsous mutant oocytes at any of the stages examined. These new data, which are presented in the corrected figure (below), further support the overall conclusion reported in the original paper that oogenesis is intact in dachsous1b mutants. As the major conclusions of the paper are not affected, the journal editors – following consultation with all authors and the Academic Affairs Committee at Albert Einstein College of Medicine – have agreed that a Correction should be provided, with an explanation of the circumstances. This course of action complies with the journal’s policy on correction of issues in the scientific record, which states: ‘Should an error appear in a published article that affects scientific meaning or author credibility but does not affect the overall results and conclusions of the paper, our policy is to publish a Correction’. We regret any confusion this may have caused.
Hat tip: Rolf Degen
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 new daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.