Online research relationships blossom in pandemic

With academic conferences on hold, novel ways of collaborating have taken off – from preprints and digital networks to AI-backed matchmaking, writes Jack Grove

August 31, 2021
two women take photos with spring blossoms
Source: Getty

Browse the full results of the World University Rankings 2022

Download a copy of the World University Rankings 2022 report


Academic conferences should, in some eyes, have disappeared a long time ago. After all, in a digital age where scholarly papers can be emailed and lectures streamed online, flying hundreds of miles to listen to a series of talks seems like a time-consuming and costly rigmarole.

But the old model of in-person scholarly events has proved surprisingly resilient, perhaps because meeting academic peers in person remains an effective way to discuss, debate and scrutinise unpublished work. As the nuclear scientist Robert Oppenheimer memorably put it: “The best way to send information is to wrap it up in a person.”

Then the pandemic hit. Huge conferences were quickly axed as international travel was largely halted, leaving scholars to find new ways to network and collaborate while confined to their homes. Various online meetings have sprung up in their place, but the biggest shift in scholarly communication has been the rise of preprints, with almost 3,900 non peer-reviewed papers posted on the bioRxiv site in May 2020, roughly 10 times the volume submitted four years earlier and almost double the number submitted a year before; overall, some 30,000 Covid-related preprints were shared in 2020 alone.

For many scientists, the opportunity to see emerging research at an early stage has been revelatory, allowing them to compare their own work against similar studies in their field. That was the case for Josh Hardy, research associate at Monash University’s Biomedicine Discovery Institute, who realised that a preprint that pinged into his inbox covered similar ground to his own team’s work on the structure of chimeric viruses. Rather than compete to see who could publish their work first, the Monash and the University of Oxford papers – which used different approaches but converged on the same unexpected result – were submitted together, an approach that helped to strengthen the robustness of each team’s findings.

“Preprints can establish a dialogue between research groups as well as highlight the uniqueness of their work – some of the research will be very similar, but there may be other findings that are different or unique to one group,” reflects Hardy on how preprints might lead to enhanced collaboration in future.

He adds that greater use of preprints will allow deeper collaboration at an earlier stage; in his case, without the preprint, the two papers would not have been compared until after publication – a process lasting months, if not years.

“While papers are being submitted and reviewed, their unique elements can be enhanced so the end result is two papers which still have overlap and independent validation but also novel components,” he says. “Even without a direct collaboration, each group can review their own paper in light of the other and re-evaluate their findings.

“Researchers who develop new tools and methodologies can also post preprints to get the word out early,” says Hardy, who explains that sharp-eyed scientists have ended up collaborating with those behind the new tools to test their effectiveness. “In some cases, the collaborative paper is published before the initial method paper is published,” he adds.

“These trends were already developing prior to Covid-19, but the pandemic certainly acted as a catalyst for the biomedical sciences as researchers have seen the opportunities and possibilities with preprints,” says Hardy.

Having this torrent of preprints flooding scientific literature creates problems as well as opportunities, says Emily Gurley, an infectious disease epidemiologist at Johns Hopkins University, who is part of the Novel Coronavirus Research Compendium, an effort by more than 50 scientists, clinicians and students associated with the Baltimore university to sift through the Covid literature to highlight original and high-quality research.

“We would often select a paper for review that was getting a lot of press, or if we saw decisions made around them, to assess the significance of their results,” says Gurley.

The short critiques of papers summarising their strengths and weaknesses would not only inform the public about their robustness but also flag whether researchers and clinicians should take them seriously, she adds. “We were able to assess whether this was a really useful method that researchers should keep up with and help clinicians understand the strengths and limitations of a study.”

This project, initially funded by billionaire Michael Bloomberg’s philanthropy, is now being supported by eLife, the non-profit open access publisher that has introduced a new system of “refereed preprints”, in which electronic papers on medRxiv are publicly reviewed by clinicians and academics.

Glowing evaluations of preprints can, however, be easily lost amid the constant stream of new papers, making it hard for researchers to spot the one that might enhance their own work, says Paul Shannon, head of technology at eLife, whose Sciety network aims to surface useful preprints for the scientific community.

“We have been seeing over 100 preprints a day, but only one of those might be important for your work. If someone you respect has highlighted someone’s work, it gives you an idea that this could be interesting for you,” he says.

“It’s happening already with Twitter, but we want to bring it together properly for the scientific community.”

With scholars able to cluster around high-quality preprints, they have met online to discuss a paper’s various merits and how they might work together on related topics, says Shannon, adding that 20 different groups of researchers were now connecting in this way.

“At the early meetings, it used to be just a handful of people. But now we’re seeing over 100 turning up,” he says of the newly formed research communities. “I wonder whether they will feel the need to return to in-person events if this digital option is there.”

Connecting with business on research activities has also been tricky for researchers during the pandemic. According to a report by the National Centre for Universities and Business (NCUB), 45 per cent of UK universities stated that their innovation-focused activities with external partners had dropped in lockdown, with alliances with automotive and aerospace sectors hardest hit, down 30 per cent to 40 per cent in both cases.

With university campus visits and networking lunches proving difficult in recent times, NCUB is hoping that a new system called Konfer will reinvigorate these partnerships. Konfer provides suggestions for links between some 144,000 academics from UK universities and research institutes and more than 46,000 businesses, based on information scanned from 1.5 million research publications, grant-funded projects and case studies.

“The pandemic has highlighted the need for businesses to move quickly to find the right people – it’s not enough to rely on the same people in your contacts book,” says Joe Marshall, chief executive officer of NCUB’s AI-powered smart matching system.

This kind of matching could, in theory, disrupt the traditional model of university-industry partnership, which has often seen businesses turn to their local institution for R&D-related input, he says. “It’s about democratising opportunity, but also extending the reach of industry into academia by locating people that they did not know existed,” he adds.

The project might even lead to new streams of public or industry funding as novel patterns of university-private funding become more apparent, Marshall suggests.

“It provides a lot of analytics, so you can see emerging technologies appearing. And as those new concepts emerge, we hope to work with public and commercial funders in these areas,” he says.

The move to largely online research collaborations has, of course, not been entirely without pain.

“I was unable for a while to see my postdoc or visit workshops in the US, which has slowed my research down”, says rocket engineer Mike McCulloch, lecturer in geomatics at the University of Plymouth, who adds that cancelled trips to laboratories in Europe “caused serious misunderstandings” with collaborators.

Overall, however, he is positive about the new ways to communicate online – including Zoom – with colleagues.

“It has given me a much better way to disseminate my work and interact with others,” he reflects. “I have also saved stress and 8,000 miles by not travelling.”

jack.grove@timeshighereducation.com


Research pillar

Pillar rank

World University Rankings position

Institution

Country/region

Pillar score

1

1

University of Oxford

United Kingdom

99.6

2

=5

University of Cambridge

United Kingdom

99.5

3

=2

Harvard University

United States

98.9

4

=2

California Institute of Technology

United States

96.9

5

4

Stanford University

United States

96.8

=6

7

Princeton University

United States

96.0

=6

8

University of California, Berkeley

United States

96.0

8

=16

Tsinghua University

China

95.7

9

=16

Peking University

China

94.6

10

=5

Massachusetts Institute of Technology

United States

94.4

=11

9

Yale University

United States

93.0

=11

=18

University of Toronto

Canada

93.0

13

15

ETH Zurich

Switzerland

92.4

14

=13

Johns Hopkins University

United States

90.8

=15

10

The University of Chicago

United States

90.6

=15

21

National University of Singapore

Singapore

90.6

17

=35

The University of Tokyo

Japan

90.3

18

20

University of California, Los Angeles

United States

89.8

19

11

Columbia University

United States

89.6

20

=13

University of Pennsylvania

United States

89.2

21

=18

UCL

United Kingdom

88.9

22

12

Imperial College London

United Kingdom

88.4

=23

22

Cornell University

United States

85.4

=23

=24

University of Michigan-Ann Arbor

United States

85.4

25

26

New York University

United States

83.1

26

=24

Northwestern University

United States

82.1

27

27

London School of Economics and Political Science

United Kingdom

80.8

28

28

Carnegie Mellon University

United States

80.0

29

61

Kyoto University

Japan

78.9

30

29

University of Washington

United States

78.8

31

23

Duke University

United States

78.6

32

32

LMU Munich

Germany

77.9

33

48

University of Illinois at Urbana-Champaign

United States

77.3

34

34

University of California, San Diego

United States

75.6

35

38

Technical University of Munich

Germany

74.8

36

=30

University of Edinburgh

United Kingdom

74.2

37

37

University of British Columbia

Canada

74.0

=38

=54

Seoul National University

South Korea

73.8

=38

33

University of Melbourne

Australia

73.8

40

=42

KU Leuven

Belgium

73.2

41

=30

University of Hong Kong

Hong Kong

72.2

42

84

Shanghai Jiao Tong University

China

71.9

43

47

University of Texas at Austin

United States

71.5

=44

=40

Paris Sciences et Lettres – PSL Research University Paris

France

71.2

=44

=35

King’s College London

United Kingdom

71.2

46

=75

Delft University of Technology

Netherlands

71.1

47

39

Karolinska Institute

Sweden

71.0

48

44

McGill University

Canada

70.9

49

46

Nanyang Technological University, Singapore

Singapore

70.3

=50

=158

Lomonosov Moscow State University

Russian Federation

69.9

=50

=75

Zhejiang University

China

69.9


The Times Higher Education World University Rankings 2022, which includes metrics on universities’ research environments, will be published at 00:01 BST on 2 September. The results will be exclusively revealed at the THE World Academic Summit (1-3 September), which will focus on the interrelationship between universities and the places in which they are located.

Register to continue

Why register?

  • Registration is free and only takes a moment
  • Once registered, you can read 3 articles a month
  • Sign up for our newsletter
Register
Please Login or Register to read this article.

Related articles

Sponsored