Archive for the ‘unreliable findings’ Category
The study followed 87 women for 8 weeks as they completed a regular routine of yoga, circuit training, or walking on a treadmill. “Suryanamaskar: An equivalent approach towards management of physical fitness in obese females” concludes that
All three methods were effective in weight and physical fitness management.
But a group of heath researchers at the University of Alabama at Birmingham led by David Allison failed to replicate the statistical tests on some of the data. In a recent letter to the editor, “Unsubstantiated conclusions from improper statistical design and analysis of a randomized controlled trial,” they express skepticism about the paper’s claims, and ask the journal to retract it:
Duke researcher Michael Foster and his former co-author Erin Potts-Kant are adding to their notice count with a major correction from late last year to a paper on how certain cells in mice respond to a pneumonia infection, citing “potential discrepancies in the data.”
The correction is actually a partial retraction: The note explains that parts of three figures should be discounted.
We’ve also recently unearthed multiple corrections and two retractions from the pair that we missed from earlier in 2015.
After questions about the data in the corrected paper arose, the authors were able to replicate most of the experiments in the paper, according to the note. But since the paper was published, the senior author passed away, closing her lab, so they couldn’t repeat all of the work.
Here’s the correction notice for “Mast cell TNF receptors regulate responses to Mycoplasma pneumoniae in surfactant protein A (SP-A)−/− mice,” published in The Journal of Allergy and Clinical Immunology:
A paper on how plants respond to bacteria has an invader of its own — data manipulation.
The “irregularities and inappropriate data manipulation” were found in a figure produced by the first author, Ching-Wei Chen, whose LinkedIn page lists him as a student at the National Taiwan University. The authors were unable to replicate the results in the figure, according to the note.
The authors are doing more experiments to verify the main conclusion of the 2014 paper, “The Arabidopsis Malectin-Like Leucine-Rich Repeat Receptor-Like Kinase IOS1 Associates with the Pattern Recognition Receptors FLS2 and EFR and Is Critical for Priming of Pattern-Triggered Immunity,” published in The Plant Cell.
The retraction note explains the what happened in more detail:
Authors are retracting a 2012 paper on cholesterol metabolism in zebrafish after realizing it included a case of mistaken identity in a DNA sequence crucial to some aspects of the experiment.
A postdoc misidentified the plasmid in question after failing to fully sequence it before including it in the experiment. A technician in the lab found the mistake, last author Steven Farber, a researcher at the Carnegie Institution for Science in Maryland, explained:
When the omitted region was correctly sequenced we discovered it had an error.
He told us in a phone interview what that felt like:
We were like, holy crap.
Next came months of back and forth with the journal, discussing whether to correct or retract the paper. Farber tells us the mistake, which affects two figures,
was unfortunate. Most of the paper is in fact correct.
The paper, “Visualization of Lipid Metabolism in the Zebrafish Intestine Reveals a Relationship between NPC1L1-Mediated Cholesterol Uptake and Dietary Fatty Acid,” published in Chemistry & Biology, has been cited 21 times, according to Thomson Scientific’s Web of Knowledge.
An article about how a COX-2 inhibitor (celecoxib) inhibits growth of prostate cancer in rats is being retracted after the authors were unable to provide an investigation committee at New York University with the backup they were asking for.
When the paper was published in 2003, first author Bhagavathi Narayanan worked at the Institute for Cancer Prevention in New York (also known as the American Health Foundation). But when the institute went broke the next year — thanks partly to lavish salaries and offices, as the New York Post reported — the authors claim they could no longer obtain back up for an image in the paper, once it was questioned years later by NYU, where Narayanan is now based.
Here’s the retraction note, published in Clinical Cancer Research:
Nature retracted a paper on protein structures today, six years after an investigation at the University of Alabama identified several structures that were “more likely than not falsified and/or fabricated” by one of the authors.
The paper came under scrutiny soon after it was published in 2006. A letter published in Nature that same year pointed out “physically implausible features in the structures it described.” That triggered the investigation at the University of Alabama, the result of which was published in 2009, identifying “nine publications related to the same protein structures that should be retracted from various scientific journals.” Everything was pinned on last author H.M. Krishna Murthy, who the investigation determined was “solely responsible for the fraudulent data.”
A 2009 Nature news article on the investigation declared that the “fraud is the largest ever in protein crystallography.”
We’re not sure what took Nature so long to retract the letter, titled “The structure of complement C3b provides insights into complement activation and regulation.” Here’s the note, which explains that not all the authors agreed to the retraction:
Key assertions in a paper on homosexuality have been removed from the Journal of the Islamic Medical Association of North America, in what the notice describes as a “partial retraction.”
The 2006 article “Homosexuality: An Islamic Perspective,” states that conversion therapy can be effective, and that gay people have poorer health. Those statements are among those that lack evidence, according to a note on the paper published in July. The retraction pulls those assertions, among others, and instead argues that a homosexual person should be helped to “accept his or her LGB identity,” and find a welcoming community.
The 2006 article is definitely a perspective — it states the opinion of the sole author, M. Basheer Ahmed, who has a private psychiatry practice in Texas, as to whether homosexuality is a choice. He thinks yes, though the science on the matter is fairly clear that it’s not.
But we still think it’s interesting that a journal chose to take back some of the statements contained in the article. Here’s the abstract from the “partial retraction” note:
A JAMA study on an inexpensive treatment for osteoporosis has been retracted because the first author falsified or fabricated data. We’ve been expecting this one: An investigation at Women’s College Hospital in Toronto, the bone researcher’s former workplace, had already revealed issues with the paper.
An internal memo sent to staff (available in full here) in October explained that the investigation had found “unequivocal evidence of systematic data manipulation” by Sophie Jamal, who had already resigned from her positions at WCH and the University of Toronto.
The study appeared to show that nitroglycerin ointment could have a small positive effect on bone mineral density in postmenopausal patients. It’s been cited 30 times, according to Thomson Scientific’s web of Knowledge. (That’s two more times than when we last reported on the paper: It was cited once by the investigation, and once by a paper that was published earlier in October, but had not yet been indexed.)
Researchers in Germany have retracted their 2011 article in the Journal of Bacteriology after another lab pointed out a fatal error in the paper.
The article, “Escherichia coli Exports Cyclic AMP via TolC,” came from a group at Tübingen University led by Klaus Hantke. The paper focuses on the crucial role of the membrane channel TolC in exporting cyclic AMP (cAMP)-cAMP receptor protein (CRP) complex, which regulates nearly 200 E. coli genes. According to the abstract:
The data demonstrate that export of cAMP via TolC is a most efficient way of E. coli to lower high concentrations of cAMP in the cell and maintain its sensitivity in changing metabolic environments.
If you could help reduce the waste of tens of billions of dollars per year in research spending, you’d do it, right?
This is the second in a series of two guest posts about the havoc misidentified cell lines can wreak on research, from Leonard P. Freedman, president of the Global Biological Standards Institute. Freedman who published a paper last summer detailing the financial costs of non-reproducible research — namely, tens of billions of dollars per year. Some of that non-reproducible research is due to the use of contaminated or misidentified cell lines. He writes about one key step to tackling the problem: Ask every scientist to use a relatively inexpensive technique to validate the identity of their cell lines.
Meanwhile, we have to deal with the issue of all the previously published papers that relied on problematic cell lines, now contaminating the scientific literature. Scroll down to the bottom of the post to take a poll on what you think should be done about those papers.
As new frontiers of science emerge, from Pluto to proteins, the very cornerstone of the scientific process—reproducibility—has also reared its head as a huge problem. Estimates of irreproducibility rates of published peer-reviewed papers range from 51% to 89%. An analysis that two colleagues and I recently published in PLOS Biology suggests the U.S. spends $28 billion per year on non-reproducible preclinical research; global spending could be up to $60 billion per year. This lack of reproducibility typically results from cumulative errors or flaws in one or more of the following areas: biological reagents and reference materials, study design, laboratory protocols, and data analysis and reporting. Given the size, scale, and especially the complexity of reproducing preclinical research, there is no single magic bullet fix. This is a difficult issue for scientists to own up to, and for the public to grasp.
However, an approach that has demonstrably addressed similar challenges in other complex and evolving industries, such as those involved in the founding of the Internet, is the expanded use of community-based voluntary standards and best practices. And here’s where we start: Read the rest of this entry »