As health authorities isolate the sick, set up quarantines and close borders, early efforts to stop a coronavirus pandemic are also trying something completely different: “open science.”
The COVID-19 outbreak that began in Wuhan, China, has become an important test case for the risks and benefits of open science — a movement long-simmering within the global research community. By calling for rapid, free and public posting of scientific findings, it amounts to a rebellion against the old ways of conducting and reporting research.
Scientists throughout the world are publicly sharing brief clinical reports and gene sequences of the novel coronavirus, bypassing the careful curation and peer review that still dominates the distribution of most scientific information.
The benefits? Lightning-fast collaborations that might help forestall a pandemic. The risks? Bad science can share that spotlight and, when something as scary as this coronavirus comes along, contribute to the cacophony of conspiracy theorists on the internet.
Overhyped headlines, snarled statistics lead readers astray
Read our special series of articles with tips and tools to help you better understand health risks, scientific research, statistics and clinical studies
In this case, because of the seriousness of the epidemic, virus genomes are being released three to six days after sample collection. "We haven’t seen this before,” said Dr. Trevor Bedford, a computational biologist at Fred Hutchinson Cancer Research Center in Seattle.
He is tracking the virus on Nextstrain, a public website he developed for scientists that provides informative and animated maps of viral evolution. The genetic sequences are pulled from GISAID, an online repository where labs from all over the world post new genomic data. Bedford's phylogenetic charts — family trees for viruses — can help guide the public health response as an epidemic unfolds, whether it be avian influenza, Ebola, coronavirus or Zika.
“I’ve been talking about ‘real-time’ phylogenetics for a long time, and now we are there,” he said.
In a blog post from his Fred Hutch lab, Bedford explained how the analysis of genomic data led him to conclude that a lack of genetic diversity in strains as the outbreak grew pointed to “sustained human-to-human” transmission, springing from a recent jump from animals to people just a month or two earlier.
“As this became clear to me, I spent the week of Jan. 20 alerting every public health official I know,” he wrote.
Bedford is an early user of bioRxiv (pronounced Bio-Archive), an online server where scientists post the first drafts of papers they may or may not submit later to academic journals. Online, these “preprints” are subject to a kind of crowd-sourced peer review. The work is disseminated to colleagues around the world, who provide valuable public critiques. By their nature, however, they are works in progress, and should not be confused with established fact.
The downside became apparent on Jan. 31, after a group of researchers at the Indian Institutes of Technology posted on bioRxiv an analysis that purported to find an “uncanny similarity” between tiny segments of the virus' genes to sequences found in HIV — implying that new virus could be a laboratory-generated mutant. The authors posted a link to their paper on Twitter, and their claim gained attention globally.
Within hours, Bedford and other genomics experts posted fierce criticism of the bioRxiv paper. They pointed out that these same short sequences can be matched in coronaviruses previously found in wild bats and are commonplace in nature. If you searched for these short sequences in gene libraries, you’ll spot them “throughout the tree of life,” Bedford said later in an interview.
The chastened authors marked their paper “withdrawn” two days later, but a contagion of weak science and outright falsehoods had been unleashed. Freely available to anyone, the story morphed and blossomed on the websites of conspiracy theorists who headlined that “Scientists Confirm” the virus was “Man-Made.”
“In the scientific realm, things corrected themselves very quickly. That’s exactly how you want science to happen,” Bedford said. “But the post was picked up by various websites that are outside of the scientific dialog, and it doesn’t seem to be easily squashed.”
Dr. John Inglis, executive director of Cold Spring Harbor Laboratory Press, co-founded bioRxiv in 2013 and manages it today. Although the site uses a screening process for bogus claims before putting up papers, he said it was “unfortunate” that this paper got through. “We have since had a great deal of discussion on how to ensure that a similar incident does not happen,” Inglis said.
When scientists began sequencing genomes of the virus responsible for the outbreak in Wuhan, China, they called it 2019-nCoV, for the "novel" coronavirus first isolated in December 2019.
The designation was widely used in the research community until Feb. 11, when the International Committee on Taxonomy of Viruses, responsible for naming viruses, declared it to be “a sister” to coronaviruses responsible for SARS, the 2003 epidemic that killed 774 in 17 countries.
They designated the new virus SARS-CoV-2, and posted the name change on bioRxiv.
BioRxiv has since added a bright yellow band across all its postings to stress that these papers are “preliminary reports that have not been peer-reviewed … and should not be reported in news media as established information.”
Aided by social media, data and scientific chatter are circulating faster and more freely, for better or for worse. In the worst cases, unproven claims can be hijacked by traffickers in fear, sensation and fraud.
Inglis said responsible journalists rely on a network of trusted, expert sources — just as scientists do among their own colleagues — to assess the veracity and importance of preprint research.
Karl Bates, director of research communications for Duke University, notes that preprints and crowd-sourced peer review got a foothold decades ago in physics and astronomy. BioRxiv is modeled after arXiv, an open-access server founded for physicists in 1991, with more than 1.6 million postings.
“I think open science is healthy,” he said. “It would be even better if the science news media did a better job of explaining that this is part of a process, that science is a sort of ongoing argument.”
As the coronavirus outbreak puts open science under the microscope, the sharp divisions over how research is conducted and distributed are coming into focus.
Dr. Sue Biggins, senior vice president and director of the Basic Sciences Division at Fred Hutch, sees a certain inevitability in the rise of open science. “We have to think about a world where journals might just go away,” she said.
“Everything’s going to be on the internet. I want to see us focus on how to make things work for the next generation, and not just where we are now,” said Biggins.
Driving the movement for open science is a mounting resentment against the power of scientific publishing houses that charge high fees for access to the works they publish, and the lengthy and opaque peer-review processes they employ. With trained scientific editors and networks of academic volunteers who analyze submissions, the premier houses produce elite journals such as Cell, Science and Nature, whose selections for print can make or break academic careers.
“If you want to fly high with the so-called ‘high-impact’ journals, the review process can drag along for a year,” said Dr. Randy Schekman, a University of California, Berkeley cell biologist who won the Nobel Prize in Physiology or Medicine in 2013.
In 2011, he was named the founding editor-in-chief of eLife, which provides free access to papers that are published after a streamlined peer-review process, and he served there for seven years. Coincidentally, eLife recently launched a pilot project, Transparent Review in Preprints, by which its peer-review services will be offered to authors of bioRxiv papers.
Schekman also is a leading advocate for the University of California library system’s decision last year to break with Elsevier, publisher of Cell and nearly 3,000 other journals, over open access and subscription fees. More than 30 prominent UC faculty members have resigned from Cell Press editorial boards. Negotiations between the publisher and the 10-campus system are expected to resume.
“Being a public institution, they wanted more for their money. They want all the work published by UC scholars to be open access, and Elsevier balked at that. It would put a huge dent in their excessive profit margins,” Schekman said.
— Dr. Randy Schekman, UC Berkeley cell biologist and Nobel laureate
As that battle carries on, the Trump administration has floated a proposed executive order that all federally funded research be made immediately available to the public upon publication — directly challenging paywalls by some publishers that restrict such access, for up to a year, to those who pay. Schekman and 20 other Nobel laureates signed a letter to Trump in support of it. “The old model of subscription publication is not a good fit for the modern electronic era,” they wrote.
The White House proposal has further alarmed publishers, particularly those run by scientific societies, which rely on revenues from subscribers to validate the science through peer review and distribute their findings.
Fred Hutch transplant physician Dr. Stephanie Lee, who holds the David and Patricia Giuliani/Oliver Press Endowed Chair in Cancer Research and is president of the American Society of Hematology, responded for the society in a letter to President Trump. She wrote that the proposal could significantly harm “scientific rigor, discovery and innovation” by threatening the resources needed to staff a high-quality peer review, curation and dissemination system. (At this time, Fred Hutch has not taken a position on the proposed executive order.)
The push for open access, however, has strong support from powerful players. bioRxiv has received funding from the Chan Zuckerberg Initiative, a limited liability company founded by Facebook founder Mark Zuckerberg and his wife, Dr. Pricilla Chan. The Bill & Melinda Gates Foundation since 2017 requires all peer-reviewed research funded by the organization to be immediately and freely available through open access publication. eLife was launched with support from the Howard Hughes Medical Institute, the Max Planck Society and the Wellcome Trust.
Stakeholders in the debate over open science have set aside some their differences in response to the coronavirus crisis. Nearly 100 publishers, research foundations and drug makers have pledged to provide open access to new research regarding the disease and share laboratory findings with the World Health Organization, “at least for the duration of the outbreak.”
Top-tier medical journals have been establishing open-access, peer-reviewed sister publications, such as the American Medical Association’s JAMA Network Open, launched in 2018. Researchers pay a fee to post peer-reviewed articles on the online site, in a process designed to be quicker than other JAMA journals'.
Dr. Frederick Rivara, a professor of pediatrics at Seattle Children’s, is editor-in-chief. “I think we in the health sciences have been a little late in adopting open science,” he said. “The principle is that, by design, research conducted openly and transparently leads to better science, and it is what we are all trying to do.”
Sabin Russell is a staff writer at Fred Hutchinson Cancer Research Center. For two decades he covered medical science, global health and health care economics for the San Francisco Chronicle, and wrote extensively about infectious diseases, including HIV/AIDS. He was a Knight Science Journalism Fellow at MIT, and a freelance writer for the New York Times and Health Affairs. Reach him at email@example.com.
Are you interested in reprinting or republishing this story? Be our guest! We want to help connect people with the information they need. We just ask that you link back to the original article, preserve the author’s byline and refrain from making edits that alter the original context. Questions? Email us at firstname.lastname@example.org