Elisabeth Bik describes herself as a “super-introvert,” the kind of person prone to weeping in closets after prolonged exposure to other people. That’s not a hypothetical, either.
But Bik, a microbiologist by training, has managed to carve out a niche for herself in science—one that doesn’t require her to interact much with other people and is perfectly suited to someone whose idea of a good time is spending 12 solid hours on a weekend scrutinizing images for signs of manipulation.
What started as a hobby has become a passion. Bik estimates that she has spent roughly 5,000 hours examining papers over the past five years. In the process, she estimates, she has identified in the neighborhood of 2,000 articles with problematic images. In 2016, she and two other researchers, Ferric Fang and Arturo Casadevall, published an article in mBio reporting on 784 papers—of more than 20,000 screened—with evidence of inappropriately manipulated Western blots. Her work has prompted dozens of retractions.
Now, after years of having day jobs in academia and biotech, Bik has decided to devote herself full time to investigating image manipulation—and to do it all as an unpaid volunteer.
Casadevall, a microbiologist and immunologist at the Johns Hopkins Bloomberg School of Public Health, calls Bik “clearly gifted in her ability to spot problems with figures. I had no idea how to look at figures to spot duplications or cropping problems and working with Elisabeth was an education for me.”
A side career is born
Bik, who studied a strain of cholera that ravaged India and Bangladesh in the early 1990s for her PhD work at the University of Utrecht in the Netherlands, has always been shy, but she hasn’t always been a data cop. She stumbled into the latter role in 2013 quite by accident, while looking at a chapter in an online book. More than half the text had been plagiarized from others’ publications—including some of her own. “I thought, ‘That’s my sentence. Give it back!’” Bik recalls.
Then, while looking through a doctoral thesis for evidence of plagiarism, she noticed a Western blot with a funny blotch. A bit later in the paper, she saw the blotch again, except this time it appeared in a different position in the image, which putatively was representing a separate experiment. “It was very distinct; it had a certain direction. But when I saw the same dot in another image, it was flipped.”
When Bik spots a problem, she notifies the journal in question and politely points out the questionable image. Since the launch of PubPeer—the online forum for post-publication peer review—she also occasionally leaves comments there, as well. Bik’s methods are decidedly artisanal. She uses an old Mac (she’d love to upgrade to a newer model) with a conventional monitor and nothing but her own eyes to find suspect figures. Spotting a problem, she has been known to fall down the rabbit hole in pursuit of others.
“For the 2016 mBio paper dataset, I only compared images within papers, but I will occasionally look into multiple papers by the same authors,” Bik says. “It is much harder, though, to remember all images within multiple papers, so I might miss reused images if there are many papers and many figures.”
Casadevall says Bik’s work first met with doubts, which soon gave way to praise. “Initially, I think the scientific community reacted with surprise at the extent of the problems,” he says. “As these results sank in, there is gratitude at her efforts and there is much greater vigilance about potential problems with figures.”
We need this to be a career that people can make money and use their talents in.—Elisabeth Bik
Some of the reactions to Bik’s work have reminded her to be careful, too. She does occasionally worry that one of the authors whose work she’s criticizing will try to cause problems for her—not an unusual occurrence in the world of scientific sleuthing. Researchers whose work is questioned sometimes send the email equivalent of a carpet-bomb to the employer and funders of the critic as a way to intimidate and dissuade future critiques. They also have been known to file lawsuits. But she takes precautions, and makes sure not to make her critiques personal.
A frequent source of material for Bik’s sleuthing is PLOS ONE, which has retracted 22 papers after Bik identified problematic figures. David Knutson, a spokesman for the publication, says her work has been good for the journal, and science generally. “PLOS ONE greatly appreciates researchers like Dr. Bik who are committed to the integrity of the published literature, and who volunteer their time to critically assess published articles and notify journals of any concerns,” Knutson the tells The Scientist and Retraction Watch. “In raising issues to PLOS ONE, Dr. Bik has provided clear reports that reflect her exceptional ability to evaluate image integrity and detect issues within and across articles.”
Advocating for change
Bik, 53, moved to the United States in 2001 with her husband, an optical engineer who had taken a job in San Jose, California. She soon landed a position in a lab at Stanford University doing early work with the microbiome. She stayed there for 15 years, leaving in 2016 to join uBiome, a company that offers personalized microbiome sequencing that is now the target of an FBI investigation into the biotech company’s financial activities. (Bik, who left uBiome late last year, is not a subject of the probe.)
After several months at another biotech, which turned out not to be a good fit, Bik recently decided to take an extended break from paid work to focus on her image work. In addition to looking through more papers, she hopes to follow up on her 2016 article in mBio to find out how journals have reacted to the findings. So far, she has heard from about a third of the 40 publications that were subject to her scrutiny for that project. “I’m not happy with that response,” she says.
Bik also has taken to social media more, posting de-identified images on Twitter and asking for comments on apparent problems. “People have been blown away by that, they had no idea this was going on,” she says. And she is fielding more requests for advice than before. “People are starting to contact me privately saying, ‘I think I see a duplication here.’”
Of course, Bik is a unicorn when it comes to her hobby. Few people have the patience to sift through thousands and thousands of articles looking for tiny discrepancies in figures. Fewer still will subject their retinas to such strain for free. Bik believes science needs to throw more resources at the task.
“We need this to be a paid job at a journal,” she says. “We need more staff at universities and institutes to deal with these things. We need this to be a career that people can make money and use their talents in.”
One person with such a job is Kaoru Sakabe of the American Society for Biochemistry and Molecular Biology, which publishes the Journal of Biological Chemistry and other journals. When it comes to Bik’s work, says Sakabe, “I am really impressed. I admire her tenacity and the fact that she keeps going back to editors and trying to get these issues resolved. It’s really amazing.”