At the Francis Crick Institute, we strongly support open access, open data and open software, and we are starting to embrace open methods and open infrastructure. Shared datasets and clear methods make findings reproducible and science more trustworthy, and they let others quickly build on our results and reuse our data in new ways, making the most of our charitable funding.

A new policy reveals a gap

We were reasonably confident that we were complying with, and often leading, best practices around open data. Then, in November 2023, the UK Medical Research Council (MRC) released a new and highly detailed data management policy. As it turned out, most of the policy covered actions that were already supported and encouraged at the Crick, but there were one or two items that caught us unawares.

One of the big ones was around metadata documentation. The line in the policy was “Metadata documentation, a metadata catalogue, or personnel with relevant knowledge and expertise, that can support the understanding and use of datasets, must be available to facilitate data sharing.” At the Crick, the data sharing policy had always been to use external data repositories wherever possible. For decades now, these have been standard and respected as quality technology in biology. They are where researchers go to look for data, they house detailed metadata, they support large datasets, and they ensure that the data is preserved. But they are scattered. Data from the Crick could be found through data availability statements or by searching a large number of repositories, but without an overarching discovery mechanism, there was not an easy way to identify all of the openly available Crick data.

More than a list of datasets

A metadata catalog, as suggested in the policy, seemed the way forward. However, we did not just want to create a listing. Given the opportunity to build something new, we realized that we could improve more generally on how the Crick’s open research is discovered. At the same time, something we have always cared a great deal about is the interconnectivity of research outputs. A researcher who finds a paper they like should be able to easily find the associated dataset and code, and arguably, we can already do that where good data and code availability statements exist. But on finding a dataset or code repository, the researcher should then be able to trace back their search and find the paper describing how it was created, and that is often more challenging.

The Crick already had its own publications metadata catalog, which allowed for searching by year, research group, type of publication, journal and topic. The metadata page included a wealth of information not always available on the journal site or in the institutional repository: