The use of preprints in the life sciences offers tantalising opportunities to change the way research results are communicated and reused, and the work of ASAPbio has been key in engaging the scientific community to promote their uptake. We fully support these goals, and consequently submitted a response to the recent ASAPbio Request for Applications (RFA). In light of ASAPbio’s understandable recent decision to suspend the RFA process for four months, we are making our proposal public here, to encourage and contribute to ongoing, open discussions on these matters.
Our consortium is led by the European Bioinformatics Institute (EMBL-EBI), with collaborators in the Collaborative Knowledge Foundation, the Barcelona Supercomputer Centre, eLife and EMBO. We appreciate that not everyone interested in preprints will have time to read the full proposal, so we summarise some of the main points here.
We put in a response to the RFA because we share the excitement and enthusiasm that has emerged recently around the use of preprints in the biological sciences. The reason for our excitement is simple – alongside the rapid communication of research, we see massive potential for innovation based on preprint content. We envision that the best route to enable these goals is through a reasonable number of preprint servers and services, coordinated through the operation of agreed community standards. The standards will allow content to be federated and/or aggregated across servers, depending on the use cases. This model allows a diversity of approaches to addressing the opportunities and challenges that preprints bring.
Between us, we are developing infrastructure and services for publishing processes, article enrichment, text and data mining tools, bioinformatics, and mechanisms for data integration and discovery. But more important than our singular contributions, we are also embedded in broader researcher and developer communities that are as enthusiastic as we are about the opportunities for innovation that preprints offer. Alongside the core elements in the ASAPbio RFA, the fundamental theme of our proposal is therefore to enable those communities to engage with preprint content and contribute to moving scientific communication “beyond the PDF”.
Our proposal is to combine existing and emerging open-source software and open data infrastructure to facilitate the ingestion of preprints from any source into a community archive and then share the content in different ways. This satisfies not only the scientific imperative of rapidly discoverable research results, but also creates a platform for innovation that has the promise to make information discovery faster and more effective in the future.
In short, the central services we envisage will enable any interested party to develop “plug-in” applications that can be used – optionally and in any order – in any part of the system. Some applications might work on individual documents prior to release (for example in quality control); others might work on the collection as a whole, post release; some might be fundamental “mission-critical” steps (like document conversions); and some might be more experimental. We propose to engage the developer and text- and data-mining communities through open challenges to invent new applications based on preprints. No-one knows where the next “killer app” will come from, so we want to foster broad participation and expose these developments to the wider scientific community.
The top priority is to support the uptake of preprints by the scientific community and ensure their citability and discoverability. But in order to realise transformative developments in the future, there are necessities beyond this.
Most critical among these is the ability to reuse preprints. By this, we mean not only that the content has a license that supports reuse (the CC-BY license), but also that the content is readily available as a whole, so that would-be application developers and text-miners do not have to struggle to gather content together. Most peer-reviewed literature is still subject to access and reuse restrictions and is highly distributed – with preprints we have a unique opportunity to support unrestricted and comprehensive reuse from the outset.
Secondly, quality metadata and the consistent application of standards are essential. We care about open standards like JATS for structuring the XML of full text articles, and are open to discussion about how this may evolve to support preprints in the future. Author names with ORCIDs, machine readable data citation, correctly identified institutions and funding sources are all critical for a connected research management ecosystem. Given these building blocks, others could develop tools that reduce the repetitive reporting burden on researchers, or services and indicators to give a wider stakeholder group a better understanding of the influence and impact of research.
Finally, a governance structure that represents the interests of the community is a necessity, as services around preprints need to remain current and address evolving user needs over time.
This approach to preprints infrastructure lends itself to reuse within different disciplinary contexts, providing a basis for cross-disciplinary standards of core elements, yet allowing adaptation by those communities according to their specific scientific requirements.
Central services are a crucial part of biology today. It is hard to imagine how biology could progress without resources such as the wwPDB, or the International Nucleotide Sequence Database Collaboration. We are excited about preprints because they offer a tremendous opportunity to move science forward in parallel with these data resources, enabling integration of research outputs and knowledge discovery. We welcome comments and discussion as we move towards these shared goals, supporting science into the future.