Recently, a biostatistician sent an open letter to editors of 10 major science journals, urging them to pay more attention to common statistical problems with papers. Specifically, Romain-Daniel Gosselin, Founder and CEO of Biotelligences, which trains researchers in biostatistics, counted how many of 10 recent papers in each of the 10 journals contained two common problems: omitting the sample size used in experiments, as well as the tests used as part of the statistical analyses. (Short answer: Too many.) Below, we have reproduced his letter.
As with most court cases, this one had a long backstory: An earlier version of the paper had listed Mallon as a co-author, but was rejected by Neuron in 2011; after the authors had a dispute over the data, a different version of the manuscript was submitted to PLOS Biology in 2012, leaving Mallon off the co-author list. So Mallon had sued to have the paper retracted.
In communications with Retraction Watch and other media, as well as during depositions of the plaintiffs, Mallon referred frequently to allegations of scientific misconduct, including the fact that the first author of the PLOS Biology paper had an unrelated paper retracted in 2010 for duplicated data. However, this case was brought under the Copyright Act, which focused on the authorship dispute.
A sixth retraction has appeared for a diabetes researcher who previously sued a publisher to try to stop his papers from being retracted.
Mario Saad‘s latest retraction, in PLOS Biology, stems from inadvertent duplications, according to the authors. Though an investigation at Saad’s institution — the University of Campinas in Brazil — found no evidence of misconduct, a critic of the paper told The Scientist he does not believe that the issues with blots were inadvertent.
Previously, Saad sued the American Diabetes Association to remove four expressions of concern from his papers; they were later retracted, even though Unicamp recommended keeping three of them published.
After a journal began tagging papers that adopted open science practices — such as sharing data and materials — a few other scientists may have been nudged into doing the same.
In January 2014, Psychological Science began rewarding digital badges to authors who committed to open science practices such as sharing the data and materials. Astudy published today in PLOS Biology looks at whether publicizing such behavior helps encourage others to follow their leads.
The authors of a paper showing a “striking and unanticipated” relationship between light and temperature in regulating circadian rhythms are retracting it when the results couldn’t be replicated.
After being contacted by another group who couldn’t reproduce the data, the authors failed to, as well. They “have absolutely no explanation for the discrepancies with the original results,” according to the note in PLOS Biology.
The investigation, by the Istituto Nazionale per la Ricerca sul Cancro, found that there were multiple “figure anomalies.” According to the note:
An explanation of inadvertent error was given for some of the issues identified, while for two issues, a satisfactory explanation could not be provided.
First author Roberto Gherzi says none of his co-authors helped prepare the figures. The authors maintain that the conclusions are unaffected, but that assurance wasn’t enough for the journal. Here’s more from the lengthy retraction note, which provides some backstory on the “serious concerns” regarding the data:
A paper has been retracted from PLOS Biology for duplicating lanes and incorporating others that “came from an unrelated experiment that had already been published.”
According to the retraction notice, first author Laura Cerchia says that the mistakes came “as a consequence of incorrect incorporation of representative blots.” Cerchia — along with her supervisor, study author Vittorio de Franciscis — “apologizes for this.” None of the other authors were “involved in the preparation of these figures, and there is no concern about the results that they contributed.”
Cerchia maintains that the paper’s conclusions are still valid, but the remaining authors write that the issues undermines their confidence in the results. According to the notice, the retraction is a result of “an institutional inquiry” at the Consiglio Nazionale delle Ricerche (CNR) in Rome, where Cerchia and de Franciscis are both based.
In 2011, the two, along with several other members of Marshall’s lab at Brown University, wrote and submitted a paper to Neuron (listing Mallon as the first author), which was not accepted. Shortly after submission, “Mallon and Marshall had a falling out,” the complaint says — specifically, they “disagreed about how Ardane should be operated and about the required standards of legal and ethical conduct.” Mallon left the lab and founded his own company, Calista Therapeutics.
In 2013, Marshall and his team published a paper in PLoS Biology, “Impairment of TrkB-PSD-95 Signaling in Angelman Syndrome,”that had some passages taken almost verbatim from the Neuron submission, the complaint says, but Mallon was not included as an author. According to the lawsuit: