A professor of cell biology in Japan faked data in an influential cancer study published in Nature Neuroscience in 2019, according to an investigation by Okayama University.
The school, which released a report with its findings last month (in Japanese), found no fewer than 113 instances of fabrication as well as problems with several images in the paper, titled “Genetic manipulation of autonomic nerve fiber innervation and activity and its effect on breast cancer progression.”
The report identified Okayama’s Atsunori Kamiya as being involved in the misconduct and recommended retracting the article, which has been cited 134 times, according to Clarivate’s Web of Science. The paper has yet to be retracted, but an editor from the journal said they are looking into the case.
Kamiya also is a former director of the Department of Cardiovascular Control at the National Cerebral and Cardiovascular Center Research Institute in Suita, Japan. The center took part in Okayama’s investigation.
A Google-translated version of the report reads:
The investigative committees of both organizations judged that the maliciousness was high (large) because it consisted of fictitious experimental results over a fairly wide range and a large number of specific misconducts were recognized.
A third-party investigative committee concluded that:
The defendant was found to have failed to comply with the ethical guidelines, and it was determined that the degree of non-compliance was serious.
When asked for the paper’s underlying data, Kamiya claimed that the hard disk storing them fell and broke during the June 2018 North Osaka earthquake. The paper versions were destroyed after chemical liquids from refrigerators and shelves fell on them during the earthquake, Kamiya told investigators.
The report also notes that Kamiya claimed the problems with figures in the paper were due to a “mix-up” with images.
According to a news story in The Asahi Shimbun about the case, “The two organizations concluded that Kamiya’s fabrications were so flagrant that he lacked a sense of ethics as a scientist.” (The newspaper retracted a 2019 story about Kamiya’s work.)
The study was first flagged on PubPeer in 2021. The probe also looked into two other studies coauthored by Kamiya, but found no evidence of misconduct.
Elisa Floriddia, senior editor at Nature Neuroscience in the UK, told Retraction Watch:
We are working on this, but haven’t reached a decision (correction or retraction).
Kamiya did not respond to requests for comment.
Hat tip: Lemonstoism, author of World Fluctuation Watch
Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at [email protected].
Probably the dog also ate his homework several times, when he was in High School.
The earthquake justification is hard to believe. The data for this event, with an epicentre close to Osaka and an intensity M = 5.5, are available on the USGS site:
https://earthquake.usgs.gov/earthquakes/eventpage/us1000eu1c/executive
The interactive map shows that the quake was practically not felt in Tokyo (located 400 km from Osaka). Indeed, the estimated quake intensity in Tokyo was M = 2, which is far from sufficient to broke a hard disk. See here:
https://earthquake.usgs.gov/earthquakes/eventpage/us1000eu1c/dyfi/intensity-vs-distance
Perhaps the hard disk was in a lab located at Nagoya, since one co-author is working in this city near Osaka. However, another inconsistency is about the time frame: the Osaka quake occurred on June 18, and the article was submitted three months later, on September 15. Does that mean that the article was written, at least in part, after the data loss event? Anyway, did the authors submit the paper while knowing full well that experimental data were no longer available? This is puzzling…
A colleague commented to me that the computer was probably located in Okayama (150 km from Osaka), rather than in Tokyo. Right, however, the 2018 earthquake was also weak in this region (M = 2-3 according to USGS).
Sad to hear this news. Academia needs to be more tolerant of mistakes by researchers and incidents caused by natural disasters. Research is never achieved only by the ability of the researcher. It is truly regrettable that a potentially innocent researcher was sacrificed due to a natural disaster such as an earthquake.
No, academia should not be more tolerant, it already is much too tolerant.
You defend Kamiya’s use of a natural disaster to explain why that data (both digital and physical) was destroyed (and as Sylvain points out, several months before the submission!), but the university found that it was not a credible explanation, as no other parties could remember this happening.
Anyone doing research with data that is not backed up to at least two USB sticks and (confidentiality permitting) two cloud services should hand in their PhD immediately. I am constantly amazed at the number of senior scientists who seem unable to take the most basic precautions to avoid data loss or mitigate hardware failure, even though a funder has typically paid them many thousands of dollars to produce those data. It’s as if the last 40 years of development of ever-cheaper and more power personal computing devices just passed these highly intelligent and data-aware scientists by, as if they had been in jail the whole time or something.