by Allegations of fakery research at a leading cancer center have drawn attention to scientific integrity as amateur sleuths uncover image manipulation in published research.
The Dana-Farber Cancer Institute, an affiliate of Harvard Medical School, announced Jan. 22 that it is requesting retractions and corrections to scientific papers after the British blogger noticed problems in early January.
The blogger, Sholto David, aged 32, from Pontypridd, Wales, is a scientist-sleuth who detects cut and paste image manipulation in published scientific papers.
He’s not the only one who has a hobby of going through the pixels. Other champions of scientific integrity are keeping researchers and scientific journals at bay. They use special software, oversized computer monitors and their eagle eyes to find flipped, duplicated and stretched images.
A look at the situation at Dana-Farber as the sleuths look for sloppy errors and outright fabrications:
WHAT DOES DANA-FARBER HAVE?
In a January 2 blog post, Sholto David presented questionable images from more than 30 papers published by four Dana-Farber scientists, including CEO Laurie Glimcher and Chief Executive Officer William Hahn.
Many images appeared to have duplicate items which would have made the scientists’ findings appear stronger. The papers being examined involve laboratory research into the workings of cells. One involved bone marrow samples from human volunteers.
The blog post included problems seen by David and others previously uncovered by sleuths on PubPeer, a site that allows anonymous comments on scientific papers.
Student journalists at The Harvard Crimson covered the story on January 12, followed by reports in other news media. It was the plagiarism investigation of Harvard’s former president Claudine Gay, who resigned early this year, that brought attention to the spotlight.
HOW DANA-FARBER ANSWERED?
Dana-Farber said they were already looking at some of the problems before the blog post. As of January 22, the institute said it was in the process of requesting six retractions of published research and that 31 other papers needed corrections.
Withdrawal is serious. When a journal retracts an article the research is usually so flawed that the results are no longer reliable.
Dr. Barrett Rollins, research integrity officer at Dana-Farber, in a statement: “Following the usual practice at Dana-Farber to review any potential data error and make corrections where necessary, the institution and its some scientists have already accepted promptly and decisively. action in 97 percent of cases flagged by blogger Sholto David.”
WHO ARE THE SLEUTHS?
California microbiologist Elisabeth Bik, 57, has been drinking for ten years. Based on her work, scientific journals have retracted 1,133 articles, corrected another 1,017 and printed 153 expressions of concern, according to a spreadsheet in which she tracks what happens after she reports problems.
She has received doctorate images of bacteria, cell cultures and western blots, a laboratory technique for detecting proteins.
“Science should be about finding the truth,” Bik told the Associated Press. She published an analysis in the American Society for Microbiology in 2016: Of more than 20,000 peer-reviewed papers, almost 4% had image problems, and about half appeared to be deliberately manipulated.
Bik’s work brings in donations of about $2,300 a month from Patreon subscribers and occasional honoraria from speaking engagements. David told AP that his Patreon income recently picked up to $216 a month.
Technology has made it easier to root out image manipulation and plagiarism, said Ivan Oransky, a New York University science educator who co-founded the blog Retraction Watch. The sleuths download scientific papers and use software tools to find problems.
Others who do the investigative work remain anonymous and post their findings under pseudonyms. Together, they have “changed the equation” in scientific publication, Oransky said.
“They want science to get better and better,” Oransky said. “And they’re frustrated by how most people in academia — and certainly in publishing — aren’t interested in setting the record straight.” They are also concerned about the erosion of public confidence in science. WHAT CAUSES MISBEHAVIOR?
Bik said some mistakes could be sloppy errors where the images were labeled incorrectly or “someone took the wrong photo.”
But some images are obviously changed and the articles are duplicated or rotated or flipped. Scientists building their careers or seeking tenure face pressure to publish. Some may deliberately falsify data, knowing that the peer review process — where a journal sends a manuscript to experts for comments — is unlikely to receive a fake.
“At the end of the day, the motivation is to publish,” Oransky said. “When the images don’t match the story you’re trying to tell, you embellish them.”
WHAT HAPPENS NEXT?
Scientific journals investigate errors that are brought to their attention but usually keep their processes confidential until they take action to retract or correct.
Several magazines told the AP they were aware of the concerns raised by David’s blog post and were looking into the matter.
___
The Associated Press Health and Science Section is supported by the Howard Hughes Medical Institute Science and Media Education Group. The AP is solely responsible for all matters.
