According to the retraction notice, published in December in Acta Neuropathologica Communications, mice engineered to have a specific genetic mutation were mislabeled as the normal or wild type group.
The notice cites an investigation by the University of Florida; we asked the university for a copy of the report. The university sent us a redacted document, which a spokesperson told us was a self-report from the researchers regarding the mislabeling. The spokesperson explained:
No formal misconduct investigation was ever conducted.
The 2016 paper analyzed what the authors believed were mice carrying the wild type form of the gene. After it was published, they began studying the other group of mice, which the authors thought had the mutated form of the gene. Last August, a student noticed that relatively few of these mice were developing motor abnormalities, whereas the mice without the mutation had exhibited “profound muscle pathology.” The authors decided to check both groups of mice to verify which had the mutation, and on September 18, they confirmed the labeling mistake.
According to the self-report, that same day, the authors informed the journal of the error. The editor suggested that the authors retract the paper, but submit a new manuscript that “set the record straight” and described data for both groups of mice.
The authors are retracting this article. The article describes mice expressing wild-type human MATR3. However, since publication the authors have become aware that all of the lines of mice described express human MATR3 containing the F115C mutation. Transgenic mice expressing wild-type and mutant Matrin were created simultaneously in their laboratory and, at a crucial stage of generating the DNA for embryo injection, as confirmed by an investigation by the University of Florida, the DNA preparations were accidentally mislabelled. All of the founders created were mosaic, requiring extensive breeding to isolate stable lines. Mice mislabelled as expressing wild-type MATR3 were the first to produce lines that stably transmitted the transgene and thus were the first to be characterized. However, as lines of mice that were mislabelled as expressing the mutant (F115C) MATR3 were ultimately established, the data began to suggest that an error had been made. Sequence analysis of amplified tail DNA from mice descended from the lines reported in the article have revealed that they express the F115C variant. The data described are therefore an accurate description of the pathology of mice that express the F115C variant of MATR3, but not of mice expressing wild-type MATR3. The authors are preparing a new manuscript reporting data from both mice expressing the F115C variant of MATR3 and mice expressing wild-type MATR3.
The paper has been cited five times, according to Clarivate Analytics’ Web of Science.
Corresponding author Jada Lewis at the University of Florida in Gainesville referred us to the self-report, which UFL sent to us.
According to the self-report, “part of the impetus to publish was pressure to support potential grant applications to fund further characterization” of the mouse model. Without that additional funding, “we would have never been able to catch this mistake.”
The authors also write that they “should have confirmed the identity” of the mouse lines at the outset:
Our mistake was trusting our staff a little too much, and we should have been more rigorous in verifying. We have learned a painful lesson in that even a very experienced technician can make a mistake.
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 daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here. If you have comments or feedback, you can reach us at email@example.com.