The coronavirus Covid-19 crisis is leading to a “sea change” in the way that researchers are collating and analysing research in a bid to keep up with the “phenomenal” growth in scholarship on the topic, experts have suggested.
According to one search portal for coronavirus research, as of 3 April more than 6,000 papers, including preprints, have been published on the topic and related areas since the beginning of the year.
Daniel Hook, head of Digital Science, which has created the portal through a search query on its Dimensions research database, compared the growth of research related to Covid-19 to the rapidly emerging artificial intelligence field of deep learning, which had taken six years to grow to a similar level.
“It is a fairly serious body of work already produced by the community,” he said, adding that a significant chunk of scholarship was also being produced in non-medical areas such as the social sciences.
He added that the fact that many publishers were making Covid-19 research open access also meant that scholars could get around the overwhelming nature of dealing with such a vast amount of information by using sophisticated search techniques such as text mining.
Open access to full article texts “allows researchers to do data mining and text mining to find things that are of relevance, so we are seeing a sea change in culture here as well [with] people working in different ways”.
He added that researchers were also adept at filtering to their own subfield to find what was relevant to them. At first, looking at the overall body of Covid-19 research may be “a bit like drinking from a fire hydrant, but I think people will be able to locate the important and relevant stuff for them”, he said.
Erica Bickerton, head of the Coronavirus Group at The Pirbright Institute, one of the UK’s leading research institutes studying the virus, also said that the scientific community was quickly adapting to ways of sifting through such a well of information.
This included public engagement bodies such as the Science Media Centre (SMC) collecting and filtering studies so that researchers could quickly access what was relevant.
“The volume of data being generated in response to the pandemic is phenomenal and could easily become overwhelming,” she said. “However, the research community is doing a great job at communicating the data around the…outbreak.
“There are lots of useful discussions going on, especially on Twitter. Journals are also collecting together their coronavirus research so it can be found easily, and the SMC is sending out relevant preprints, which is helping me to keep track of the data being publicly released.”
Keeping on top of which preprints – which are not peer reviewed − are relevant and have robust methodologies is one of the key challenges emerging from the scientific response to Covid-19.
This could be a challenge that increases over time. According to a snapshot of Dimensions data from 3 April, about 17 per cent of the 6,659 pieces of research published this year on the topic were preprints.
Dr Hook said that this seemed to be a growing proportion over the period, although much of this could be down to a lot of the peer-reviewed research published in January or February coming from projects in related areas started last year.
“It is surprising that 1,135 [of 6,659] are preprints because you would think everyone is putting their research” out quickly in this format, he said. But he added that “you are seeing people changing their behaviours rapidly…I think if you looked at that on a day-to-day basis you would see the proportion of preprints increase.”