Victoria Yan started her preprint advocate journey as the Project Coordinator for ASAPbio’s ReimagineReview. She is now an Open Science Research Information Specialist at EMBL in Heidelberg, Germany. In the following interview, we discuss how EMBL is promoting the use of recognition of preprints from an institutional perspective.
How do preprints fit into the bigger picture of open science at EMBL?
Preprints are relevant for all research articles at EMBL. Preprinting became a requirement in EMBL’s Open Science Policy on December 21, 2021. For EMBL, preprints can enable our researchers to maximize the impact of their research.
As a signatory of DORA, EMBL also aims to improve transparency and fairness in research assessment. Thus, the research assessment section of our policy also states that research outputs such as preprints, peer review contributions, data and software should be considered alongside publication. Preprints are one of our priorities in our open science roadmap in both how we disseminate research and how we diversify the types of research outputs in research assessment.
What’s necessary to create the change you envision?
We needed to communicate the preprint requirement and Open Science Policy change through various channels. In addition to EMBL-wide communication, we identified opportunities to integrate open science training and introductions to concepts like preprinting. Some examples where we have integrated open science training include Ph.D. and postdoc fellow courses, faculty retreats, software workshops, newcomer introductions etc. In addition, we include a check for preprints in our open-access publishing workflows for EMBL lead author articles. Lastly, inspired by a strategy from ASAPbio, we found our local preprint advocates like ASAPbio Board Member/EMBL Group Leader Gautam Dey and EMBO Director Fiona Watt, to share their perspectives on the importance of preprinting.
What barriers are in the way of that change?
Since the introduction of the preprint requirement, 76% of all primary research articles (EMBL lead author articles ~15% of all articles) processed in our open access workflow were posted as a preprint. In many cases where the researcher did not publish the preprint, it was due to insufficient time and resources. Preprints integrated as a part of the publication process can lighten the workload for researchers.
In our central open-access publication workflow, EMBL authors contact the Open Science team upon the acceptance of their article. We support the authors in checking licensing, payment of the Article Processing Fee, and deposition of their article as full-text in our institutional repository EuropePMC. In the workflow, we incorporate a check of compliance with our Open Science policy, where we remind authors to post a preprint. At this stage, the article has already been peer-reviewed and accepted, therefore some of the benefits of early dissemination from earlier preprinting are diminished. In some cases, authors thought that since the article would be published open-access, then posting a preprint is not necessary. Thus, we have more work to do to convey the advantages of preprinting and help researchers find their own motivation to post preprints.
What approach have you been using to identify the preprints created by EMBL researchers?
We have an ongoing project to track EMBL preprints, as a part of a larger effort to monitor open science at EMBL. We are interested in tracking our progress in implementing the Open Science Policy and recognizing preprints as an important research output at EMBL.
Our starting point is peer-reviewed publications. We track our publications using preprint linking on Europe PMC. We query EMBL’s publications, and whether there is a preprint link. An alternative strategy is to first assign our institutional IDs ROR ID to our known publications on Europe PMC (these links are not publicly displayed, but help us to track publications using the Europe PMC database). This allows EMBL publications to be searchable with the ROR IDs. With a ROR ID query, we would be able to show how many publications have been posted as a preprint, how many have data published etc.
Another possibility is to identify EMBL preprints using researcher identifiers ORCIDs. We run an annual ORCID campaign across the organization, inviting all researchers to create ORCIDs, and also keep their profiles up to date, by claiming their preprints using the EuropePMC ORCID claiming tool. Our campaigns included messages from EMBL’s leadership and webinars, followed by targeted emails, which allowed us to achieve near 100% ORCID adoption at EMBL. With high levels of ORCID adoption, we can obtain a more complete and comprehensive source of information on preprints and research outputs from EMBL authors.
What challenges do you face in identifying preprints?
Looking for preprints from publications is a limiting factor. This means that we are only able to find preprints at a much later stage. We found that ~50% of our known EMBL preprints are not linked to the published journal article. In some cases, the preprint exists independently of any journal publications. Lastly, the variety of preprint servers used also can pose a challenge, due to a lack of indexing and linking.
Moving forward, having a strong adoption of ORCID at EMBL will help us discover our preprints. I’m very curious how other research organizations and universities are currently tracking their preprints, and hope those who are interested reach out to me! Please reach out by email at victoria.yan@embl.de.
What strategies have been effective in changing attitudes, policies, and practices?
I think what works well is that researchers can see that EMBL has made open science and open scholarship a part of the core mission. The Open Science Policy made preprinting relevant for all EMBL researchers.
High-level engagement and support from EMBL’s leadership made the Open Science Policy possible. A broad and non-exclusive group led the development of the Policy. The group first surveyed current open science related practices, which also advertised the idea that the Open Science Policy was in development. The survey results showed that there were many good practices in place, however, a policy would formalize these practices across the entire research organization. The working group developed a separate implementation guideline to contain the practical aspects of how to adopt the Open Science policy. The group that developed the implementation guidelines is continually working to improve user-friendliness and keep the guidelines up-to-date.
Adoption of the policy wouldn’t be possible without training our community of researchers. Therefore, we began to coordinate open science training across EMBL. You can see a catalogue of our publicly available training resources here. At the moment, we have mandatory open science training for all PhD students, 75% of our post-docs, and all of our principal investigators. To make clear that openness is a norm at EMBL, we continue to recognize and highlight open research projects and leaders.
