Papers Based on Misidentified Cell Lines Top 32,000

An analysis of contaminated literature finds that tens of thousands of papers used cell lines of questionable origins—and these were in turn cited by hundreds of thousands of other papers.

Kerry Grens
Oct 16, 2017

WIKIMEDIA, KAIBARA87Cell line misidentification is rampant throughout biomedical research, and a new analysis quantifies its impact on the scientific literature, finding more than 32,000 papers used lines with no known original stock. “In this case, it must be assumed that all primary literature could be based on false grounds and should at least be treated with caution,” the authors write in their report, published in PLOS ONE October 12.

The authors based their study on a list 451 cell lines flagged by the International Cell Line Authentication Committee as not having authenticated stock, meaning they are likely mislabeled. They then went through the Web of Science literature database to grab papers based on these lines. “As we only searched for cell lines known to be misidentified, this constitutes a conservative estimate of the scale of contamination in the primary literature,” they wrote.

See “The Great Big Clean-Up”...

The team found that 46 of the contaminated studies had been cited more than 1,000 times, and another 2,600 had been cited more than 100 times. In total, around 500,000 subsequent papers referenced the 32,000 studies based on misidentified cell lines.

In a press release, the authors offer a potential solution: flag each of these papers with a disclaimer about the cell line. “It would then be up to readers to decide whether it's a problem or not, because sometimes it really doesn't matter. Basically, we want to caution people to be careful with the interpretation of results.”

See “The Zombie Literature