A genetics researcher came across an interesting paper earlier this year on the gene he studies. The scientist, a doctoral candidate who asked not to be named, decided to take a closer look at which part of the gene, SNHG14, the authors targeted to measure its expression. He ran the sequence of the short strand of DNA, called a primer, given in the paper through a database and found the sequence matched with a completely different gene.
The scientist searched through similar papers and found 19 more across as many journals with the same problem: all their “SNHG14” primers matched with the gene MALAT1/TALAM1. There may be more, but he stopped looking.
Two of the papers he found have been retracted. One appeared in 2023 in Computational and Mathematical Methods in Medicine, a journal Wiley acquired from Hindawi that is no longer publishing. The notice cites inappropriate citations and peer review manipulation. The other article, published in 2022 in the International Journal of Oncology, was retracted for plagiarism.
The list includes articles in journals from publishers Wiley and Elsevier. Editors for Leukemia and Lymphoma, a Taylor & Francis title, told Retraction Watch in an email they were investigating the matter, and an editor for European Review for Medical and Pharmacological Sciences, published by Verduci, said one of its papers was already under investigation for “potential duplication or manipulation.” The editorial office of American Journal of Cancer Research, an e-Century Publishing title, said it would retract its offending paper if an investigation finds the primers were incorrect.
The researchers of all 20 papers list affiliations in China.
Jennifer Byrne, a molecular oncology professor at the University of Sydney and the director of biobanking with NSW Health Pathology in Australia, has researched the issue of incorrect primers extensively.
Byrne called incorrect primers “canaries in a coal mine.” The errors may be small, but they can indicate a larger problem, such as paper mills at work or other types of fraud.
Although Byrne hasn’t studied SNHG14 or the MALAT1/TALAM1 primers, we wanted to know: What are the odds 20 papers describing the same gene independently used the same, incorrect primer?
Since the human genome is 3 billion base pairs long, “the likelihood of people using the exact same sequences wrongly is really quite remote,” Byrne said. “There’s an unlimited number of wrong answers for any question, only one right one. So why would somebody come up with the wrong answer over and over and over again if there wasn’t some kind of link?”
One explanation is the papers are products of paper mills. “We think that paper mills just get nucleotide sequences wrong, because individual papers often describe a lot of different sequences per paper,” she said. The sequences could also be evidence of individual misconduct “where people just took the data out of another paper and reused it.”
In her research, Byrne has even found cell lines with identifiers that don’t actually exist, calling into question whether the error was a typo or misconduct. “The conclusion is they didn’t do at least some of that work, and possibly they didn’t do any of it,” she said. “It’s sickening.”
While she has attempted to get these papers corrected or retracted, few journals respond: Her team has contacted journal editors about 420 papers with contaminated and/or nonexistent cell lines, but only 12 articles have been retracted to date, Byrne said.
“I think it’s because with sequences, a lot of people just don’t understand what they mean,” she said.
Byrne has continued this work and is now cataloging errors in specific cell lines and primers. Her team is also looking at which journals might be targets of paper mills and what makes them more susceptible.
Human genome research is especially vulnerable to paper mills because the results are easy to fabricate and create “many individual manuscripts at scale,” Byrne and her team wrote in a 2022 review. “By approaching the gene research literature with a more critical mindset, researchers can avoid the costly, time-wasting and misleading traps of targeted gene research produced without experiments.”
Update: This story was updated on March 27, 2025, with minor clarifications to Byrne’s research.
Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on X or Bluesky, like us on Facebook, follow us on LinkedIn, 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].