‘Data available on request’ used to be one of the most ubiquitous phrases in research manuscripts. It signalled insouciant acquiescence with the expectations of transparent and collaborative science, but little intention to proactively do anything about it. Fortunately, the phrase is becoming rare as open science advocates have succeeded in shifting the culture of data sharing. Journals, including Nature Ecology & Evolution, have also played a part in achieving this by working with authors to provide better ways of sharing their data.

There are occasionally legitimate reasons why data should be restricted, such as privacy, sensitive location information, Indigenous data sovereignty or proprietary restrictions, but the vast majority of data can be shared in a broadly accessible manner. So, although we do allow ‘data available on request’ in certain circumstances (accompanied by an explanation), for the past five years we have been asking all other authors to share their data up front with the published article and during review. The best way to do this, which we strongly encourage, is in a recognized data repository. Compared with sharing data within the article itself (often in the supplementary files), repositories are more compliant with the requirements for data to be FAIR (findable, accessible, interoperable and reusable). Data in a repository are more likely to be found and reused by other researchers, both for the purposes of replication and for other research projects.

For some data types (such as sequence data), there are specific mandated repositories with formats that are structured for the relevant data type. However, more frequently for ecological and evolutionary data, a generalist repository is suitable and authors therefore have a choice of which to use. Details of both types of repository can be found on Springer Nature’s author guidance pages.

One of the most striking conclusions of the latest State of Open Data report is that three quarters of researchers feel they have received no support with planning, managing and sharing their research data. Few researchers are experts in data science itself, and most researchers have many other demands on their time. There is therefore a gap between what researchers may wish to do in terms of open data, and what they actually feel able to do. Closing this gap requires action by institutions, funders and publishers, and is one of the reasons that we have partnered with figshare to provide integrated data and manuscript submission.

Since 2022, Nature Ecology & Evolution has offered authors the option of uploading their data to figshare as part of the initial submission process. This free service gives authors a citable data link that remains private while the manuscript is under consideration but is automatically published on publication of the associated research article, which allows smooth compliance with journal and funder data-sharing mandates. For manuscripts that are not accepted for publication, the authors have the option of removing the figshare file. The figshare submissions are checked for technical and ethical compliance, and the authors retain full ownership of their data. One of the other findings of the State of Open Data report is that authors want proper credit for their data sharing, and one step towards this is the fact that the figshare submission has its own DOI.

As well as the benefits to authors of data-sharing compliance on publication, integrated figshare submissions allow reviewers easy access to the data while the manuscript is under consideration. This is important as the checking of the underlying data is a key part of the peer review process, and any hurdles to this can cause delays and additional rounds of peer review that inconvenience both reviewers and authors. At Nature Ecology & Evolution, we check that data are made available at initial submission for all manuscripts, regardless of whether they use the figshare integration. However, it is still fairly common for reviewers to identify missing elements that were not apparent to editors. We encourage authors to check carefully that all relevant data are available and easily accessible to peer reviewers, and to use the figshare integration where possible.

Incomplete data submission causes, at best, a delay as the data are obtained, but often it necessitates a full additional round of peer review if the omission is only noted in the formal peer review report. For this reason, we do ask reviewers to let the editor know as soon as they notice an omission, rather than waiting until they submit their review. The editors are happy to field queries if a reviewer has any doubt about what should be available to them. Lastly, for many submissions, making the code reproducible and available upfront is equally important for smooth and comprehensive peer review. Nature Ecology & Evolution has joined other Nature journals in integrating code submission with Code Ocean1, and authors also have the option of sharing their code and associated documentation through repositories such as GitHub and Zenodo.

The shift to better data sharing is beneficial to the entire scientific community as a public good, and we as a journal will continue to develop ways of facilitating this. However, it is worth noting that sharing data in a repository has also been shown to benefit the authors directly in terms of increased citations of the research article2.