The COVID-19 pandemic has been called a “stress test” for open science — one that, by most accounts, it passed. After all, the sharing of new scientific knowledge and data about the SARS-CoV-2 virus and its impacts has been fast and furious.
But not all experts are convinced that open will become the new norm for science in Canada. At least, not yet.
“I was more optimistic a year ago than I am today,” said Vincent Larivière, the Canada Research Chair on the Transformations of Scholarly Communication at Université de Montréal, in an interview with Research Money.
Over the past 18 months or so, Larivière and his research team have analyzed how open science practices have changed, specifically, whether the research community has embraced open-access publishing, open data and the adoption of preprints. The results, which he presented at the United Nation’s Open Science Conference in August, were underwhelming. Research papers and data on COVID-19 were not particularly accessible, and submissions to preprint servers — the most rapid way of disseminating new research results — had also declined.
Aled Edwards, founder and Chief Executive of the Structural Genomics Consortium and a longtime proponent of open science, said that while there are stories of people sharing “here and there,” many of the biggest scientific contributions of the pandemic have not been open, such as vaccines and diagnostic tests.
“Open science is not the end, open science is the means to an end. And if you look at the ends and track backward and ask how they were achieved, very few of them have been through open sharing,” he told Research Money.
But Canada’s Chief Science Advisor, Mona Nemer, who signed a joint statement last year calling for all COVID-19 publications to be made open upon publication, is more sanguine. In a statement to Research Money, she cited the early sharing of the coronavirus genome, the accessibility of coronavirus research in international journals and the production of cross-disciplinary and accessible knowledge products by grassroots organizations of researchers like CanCOVID as examples of the role open science has played in the pandemic.
“The pandemic has shone a light on the imperative of open science, and future challenges will benefit from the new practices, attitudes and mindsets that are now emerging,” Nemer said.
Making public health data sharable
One field in which open science practices have made an impact during the pandemic is public health, where COVID-19 data is increasingly shared internationally, nationally and locally.
In Spring 2020, Genome Canada launched the Canadian COVID Genomics Network (CanCoGEN) — an initiative to expand genomic sequencing efforts and make genomic data sharable. Laboratories across the country sequence viral samples from people testing positive for COVID-19 and share them with public health authorities. To date, CanCoGEN's VirusSeq initiative has sequenced more than 150,000 viral genomes.
“Each province has been able to use that information very proactively to inform public health and policy decisions at a provincial level. I think that's a real success story,” said CanCoGEN’s director, Catalina Lopez-Correa, in an interview with Research Money.
CanCoGEN’s data are also shared nationally and internationally. Prior to the COVID-19 pandemic, sharing this kind of public health data across provincial and territorial boundaries in Canada was not the norm. CanCoGEN has been pushing to change that.
Earlier this year, it established the open-source and open-access Canadian VirusSeq Data Portal, which hosts all Canadian SARS-CoV-2 sequences and related health data. But even with the portal, Lopez-Correa said there are numerous challenges to data sharing nationally, ranging from the technical and cultural to logistical and political.
To understand the role of human genetics in COVID-19, CanCoGEN is also sequencing the genomes of Canadians infected with SARS-CoV-2 and making those results available to the research community. At the end of the two-year initiative, called HostSeq, the organization aims to have openly published 10,000 whole genomes, which will likely be the largest open cohort in Canada.
Lopez-Correa said HostSeq could serve as a roadmap for other national health initiatives and that CanCoGEN is committed to sharing what it has learned during the pandemic response, what barriers to data sharing still exist and how those barriers can be overcome.
“I think all the pain through establishing this process of sharing for CanCoGEN is really paving the way for future data sharing in other areas in health,” she said.
Most COVID-19 data still hard to access
Toward the beginning of the pandemic, the Wellcome Trust issued a statement calling on researchers, journals and funders “to ensure that research findings and data relevant to this outbreak are shared rapidly and openly.” Prestigious journals, publishers, preprint repositories and funders, including the Canadian Institutes for Health Research (CIHR) and the Natural Sciences and Engineering Research Council of Canada, endorsed it.
But when Larivière’s team analyzed data availability statements in COVID-19-related articles submitted to preprint servers including medRxiv, they found that only a minority of data sets — approximately 11 percent — were available without restrictions. Approximately 42 percent of preprints mentioned that the data were available conditionally.
“This finding indicates — contrary to what many have suggested — that a global pandemic is not sufficient to radically modify scientific practices, and that the Wellcome Trust statement had little effect," the researchers concluded.
In an interview with Research Money, Sarah Viehbeck from CIHR said the agency included several open science requirements in its COVID-19 funding opportunities that went “over and above” the Tri-Agency Open Access Policy on Publications. One specified that data should be shared in alignment with the Wellcome Trust’s statement.
The agency intends to monitor compliance with these open science requirements, added Viehbeck, who is CIHR’s Associate Vice-President Research - Evidence Integration. “We set our parameters upfront and we’ll be looking at how those parameters were actually implemented,” she said.
Peer review is changing
Larivière’s team also found that total submissions to the preprint servers bioRxiv and medRxiv have dropped from their peak in the late spring of 2020, though submissions to medRxiv are still up from the pre-pandemic baseline. He said the change may reflect the fact that many researchers who rallied to work on COVID-19 have switched back to their core research projects and “their usual way of disseminating research findings.”
But he said the rise of preprints, even if temporary, has led to a change in the perception of the role of peer review in the research ecosystem. In the past year, some scientific journals have also changed their submission practices from “review, then publish” to “publish, then review” in recognition that peer review can help improve the quality of science but is not the sole arbiter of right and wrong knowledge.
“I wouldn't say peer review is dead. But there’s another nail in the coffin of peer review, at least as we know it,” he said.
The explosion of preprints and their place in research assessment is something that research funders like CIHR are grappling with. For example, the Australian Research Council recently rejected 20 grant applications that referenced preprints and other non-peer-reviewed research outputs, sparking an outcry in the research community which largely viewed the move as outmoded.
“We’re watching that [preprint] space quite closely, both in terms of how to assess preprints within the review process, but also their impact on the science-policy interface that has been so close in COVID-19,” said Viehbeck. “When evidence bases are emergent, how do you kind of manage that?”
An opportunity to do better
"The crisis is affecting everyone. But there are countries that were more ready than Canada to make research open," Larivière said.
“I think the crisis gives us an opportunity to do better."
At CIHR, Viehbeck said the pandemic has reinforced the agency's commitment to open science and its central place in the strategic plan, adding that the agency will be considering its policies and practices going forward.
“Truly incenting open science approaches as part of the research excellence frame is going to be critical to ensuring that Canadian health research will be internationally recognized as inclusive, collaborative, transparent and focused on real-world impact,” she said. “That's the challenge we laid forth within our strategic plan for the next 10 years.”
Aled Edwards suggested that Canada could be much bolder. As the nation looks toward rebooting the economy post-pandemic, he said open science could be part of our innovation policy, a way to attract R&D investments and spur R&D growth.
“We’ve never tried the open science way,” he said.
"Maybe there are alternative ways to work with the private sector to build a safer Canada and a safe world."