|
Lenny Teytelman can still recall
his days as a PhD student 20 years ago when he accessed public databases for
his studies in yeast genetics. “My research would be technically impossible
had researchers not deposited their data into the National Center for Biotechnology
Information,” says Teytelman, the co-founder of protocols.io, a platform for
open-access protocols, based in Berkeley, California. (Protocols.io was
acquired by Springer Nature, which publishes Nature, last July.)
Open science is a broad term that
refers to the movement of making the entire research life cycle freely
available to everyone, from citizens and students to research professionals.
This includes sharing research plans, protocols, materials, data and papers
through open-access platforms.
The practice of open science is on
an upswing. PLOS, a non-profit publisher of open-access journals, found that
the rates of data-repository use rose from 22% in 2019 to 28% in 2022 for
more than 71,000 papers published in its journals during that time. The rates
of preprints associated with published articles also increased, from 15% in
2019 to 24% in 2022. A 2006 study that analysed close to 1,500 published
papers found evidence that open-access articles had higher numbers of
citations by peers than did non-open-access articles published in the same
journal after controlling for factors such as field, the number of authors
and journal impact factor (G. Eysenbach PLoS Biol. 4, e157; 2006).
Although many researchers
wholeheartedly embrace open-access publishing, the open sharing of laboratory
materials, reagents and protocols has seen a slower adoption, mostly owing to
a lack of awareness on how to properly share them and poor incentives. A
paper published in September last year in Nature Communications found that
the majority of more than 22,000 survey participants had favourable attitudes
towards open research (J. Ferguson et al. Nature Commun. 14, 5401; 2023). Of
the participants, 90% had engaged in at least one open-science practice such
as sharing data and code. Compared with a decade ago, open-science practices
had increased from 49% in 2010 to 87% in 2020.
However, there is a disparity
between attitudes and behaviours. Across all disciplines represented in the
survey — economics, political science, psychology and sociology — the
percentage of researchers who had openly shared their data and code was much
lower than the percentage of researchers who supported the idea of open
research.
“Translating beliefs about open access into
behaviours is challenging,” says Melina Fan, co-founder and chief scientific
officer of Addgene, a non-profit platform for researchers to share DNA
experimental materials called plasmids, in Watertown, Massachusetts. But
repositories such as Addgene can play an important part in enabling resource
sharing, she says. “Repositories lower the barrier to sharing and provide the
infrastructure needed to change research culture.”
Open-science organizations
Tsuyoshi Nakagawa, a plant
geneticist at Shimane University in Matsue, Japan, found that a gene cloning
kit worked particularly well to introduce genes into plants by means of
plasmids. “As I worked in a research support centre in the university, it
felt natural for me to share my experience and materials with the community,”
he says.
“However, after I started sharing
[plasmids], I received too many requests which took time away from my work.”
To save time, he started using Addgene, which facilitated the distribution of
the plasmids. Since 2016, Nakagawa has deposited more than 80 plasmids on
Addgene.
The sharing of tangible scientific
resources, such as plasmids, requires a materials transfer agreement. When
researchers share materials such as proteins and chemicals across
institutions, the process of material transfer can take weeks or months. Open-science organizations such as Addgene help to complete all legal
paperwork pertaining to material transfer behind the scenes in a few days to
facilitate the sharing of scientific materials, including plasmids,
antibodies and viruses.
Fan says that Addgene can enhance
quality control on top of being a distributor. “When we receive plasmids from
depositors, we sequence them to ensure that the sequences match what the
depositors claim. This is important for reproducibility and transparency.” Fan
also recommends that scientists deposit their materials in global
repositories that are well established and financially self-sustaining, to
better extend the reach and longevity of materials.
Nakagawa advises scientists to
check with representatives from open-science organizations and experienced
colleagues when there are concerns. For instance, he worked with Addgene
staff to ensure that the organization could internally reproduce the plasmid
materials he wanted to share. “While funders and publishers can demand
researchers to share through policies, open-science institutions and
communities play a pivotal role to promote a research culture that normalizes
sharing practices,” says Teytelman.
The Center for Open Science, a
non-profit organization in Charlottesville, Virginia, aims to achieve exactly
that by following a ‘theory of change’ philosophy. “Our strategy consists of
five steps: making open-science practices possible, easy, normal, rewardable
and eventually a requirement,” says David Mellor, director of policy at the
centre. “The first activity in our strategy for culture change is to develop
infrastructure to enable sharing, followed by an easy-to-use ‘user interface’
that makes it easier to share.”
For example, researchers can
register their studies on the Open Science Framework registry, the centre’s
open-sourced web platform. The centre has helped to create communities for
researchers who are taking on open science, such as the Society for the
Improvement of Psychological Science, as a way for them to see open-science
practices as typical and inherent parts of research.
The centre’s staff have also
thought of ways to incentivize open science, such as gathering evidence of
sharing, which researchers can add to grant proposals, and ways for journals
to prioritize the peer review and acceptance of registered studies.
Winning over the sceptics
In addition to the view that
sharing materials adds to the workload, sceptics also criticize open-access
platforms that rely solely on the research community to self-police standards
for giving credit when due in the form of citations, authorships and
acknowledgements. The absence of active enforcement of such standards might
discourage some researchers from sharing. The use of tracking technology can
help to change the minds of those who are worried about not getting credit
for their work (see ‘Five tips on how to share laboratory materials
effectively’). For example, protocols.io has introduced metrics including the
number of article views and citations, and it gives an option for users to
vouch for the reliability of protocols.
Five tips on how to share
laboratory materials effectively
1. Seek out guidelines on sharing.
Check policies from funders,
publishers and institutions to determine whether it is mandatory or
recommended to share.
2. Decide with collaborators
whether to share from the outset.
Discuss the terms of sharing as a
research team. Although open science encompasses research protocols, raw data
and experimental materials, they do not always need to be shared to an equal
degree, for example, with domestic and international researchers.
3. Start early to minimize the
administrative pain of sharing.
Sharing takes extra effort, and to
reduce the time needed, use tools or implement practices that make sharing
easier, as early as possible. For instance, consider introducing explanatory
lines when you start coding.
4. Track who has accessed your
materials.
A big motivation for researchers
to keep sharing is when they see their colleagues using their research. As an
incentive, open-science platforms are using technologies to provide
researchers with metrics of how well received their materials are.
5. Report plans and achievements.
Some funders such as NASA and the
US National Institutes of Health now require researchers to incorporate
open-science data-management plans and evidence of sharing into grant
proposals.
Another reason that researchers
are hesitant to participate in open science is poor awareness and even
negative sentiments surrounding the movement, particularly in low- and
middle-income countries (LMICs). Magaret Sivapragasam, who used to work as a
senior lecturer and process engineer at Quest International University in
Ipoh, Malaysia, says that in her country, the concept of open science is
still new compared with North America and Europe.
“Researchers in Malaysia have the
impression that you pay to publish in open-access journals, which is
associated with predatory journals. I do not want the quality of my work to
be judged like that,” says Sivapragasam, who is now a master’s student in
science communication at the University of the West of England in Bristol,
UK. Even so, she admits that open science has benefited her work. “When I was
doing research, I was always accessing open databases to know the toxicity
levels of compounds and to cross-check my experimental data [with those] from
other labs.”
Lamis Elkheir shares a similar
experience to Sivapragasam. As a pharmaceutical chemistry lecturer at the
University of Khartoum, Sudan, and a PhD student in a joint programme between
the University of Tours in France and the Mycetoma Research Center in Khartoum,
Elkheir says that there is limited awareness of the open-science movement in
many African countries. “This leads to a scarcity of opportunities for
open-science discussions with my friends and colleagues,” she says. “However,
I believe that grassroots initiatives can change this.” Leveraging the
training she received from the open-access publisher eLife, Elkheir helped to
organize the Global Dynamics in Responsible Research virtual symposium in
December 2022, which focused on equity in research, open-science efforts in
LMICs and multilingualism in open science.
Sivapragasam suggests that to win
over sceptics in LMICs, open-science organizations should interact directly
with local researchers. “Agencies can empower researchers in developing
nations with training and resources to raise awareness. Open science is a
global effort, and no one should be left behind,” says Elkheir.
Is there a good reason not to
share?
Although sharing benefits science
in general, there might be instances when it is not appropriate to share
materials, such as when labs and biotechnology firms want to commercialize
their technology and are concerned about proprietary rights over their data
and materials.
“Although we increasingly talk
about open science, in practice, there is a spectrum of openness that
exists,” says Fan. “The terms of sharing can be legally defined to protect
commercial interests while advancing open science.”
A great example is the invention
of the CRISPR gene-editing tool. As soon as the papers related to the technology
were published, the various research groups involved deposited the plasmids
on Addgene, which saw a huge spike in demand for them. Around the globe,
thousands of researchers tested the plasmids for all sorts of applications.
“I would argue that open sharing in this instance produced data that
demonstrated to investors that the technology worked,” adds Fan. In November
last year, the world’s first CRISPR gene-editing therapy, Casgevy, was
approved in the United Kingdom to treat sickle-cell disease and
transfusion-dependent β-thalassaemia.
In research, academic reputations
are so highly valued that researchers might be overly cautious about sharing
data and methods that are not yet fully reproducible. Mellor suggests that
more education around the idea that science is a work in progress could help
to convince open-science holdouts. “I am optimistic that as we see more
researchers engaging in open science, sharing will become a norm,” he says.
“We will see the community driving the open-science movement in the future to
achieve reproducible and equitable research.” [1] | 
