What is a trusted repository?
The Horizon Europe programme, the EU's main research and innovation funding framework, mandates that research data be stored in trusted repositories. This requirement is detailed in the Annotated Grant Agreement (AGA), a binding document for researchers and research-performing organisations (RPOs). Annex 5 (pp. 373–374) of the AGA outlines the criteria for trusted repositories, emphasizing reliability, accessibility, and long-term preservation. While not all repositories must meet every criterion, they should align sufficiently with these standards. Additionally, repositories should have features that enhance functionality and usability. These features include:
Certified repositories: Certified by international organisations or government-authorized bodies (e.g., CoreTrustSeal, nestor Seal DIN31644, ISO16363).
Disciplinary or domain repositories: Commonly used and endorsed by research communities, recognized internationally.
General-purpose or institutional repositories: Must meet the following criteria:
- Ensure the integrity and authenticity of content.
- Publish policies on access, security, and long-term sustainability.
- Assign persistent unique identifiers (e.g., DOIs).
- Provide metadata fields for provenance discovery, reuse, and citation.
- Feature standardized, machine-readable metadata.
- Offer convenient licensing, especially Creative Commons;.
- Provide means for long-term preservation.
- Include mechanisms for expert curation of materials.
Selection of a repository
When selecting a repository for research data, it's crucial to consider operational criteria beyond technical and organisational aspects. These include features like computational power, visualisation tools, and simulation capabilities. Such affordances enhance the usability of datasets, enabling advanced analysis, integration with workflows, and deeper insights. For instance, visualisation tools help explore complex data, while simulation capabilities support modeling and hypothesis testing within the repository. These features add value to the data and promote its reuse, maximizing the impact of research outputs. Researchers should ensure that a repository's services meet their specific needs and those of their research communities.
Re3data.org) is a valuable resource for identifying trusted repositories, offering a comprehensive registry across disciplines. Its advanced filtering options allow users to search based on criteria like controlled access, certification standards (e.g., CoreTrustSeal), or supported data types. This makes it easier for researchers to find repositories that meet funder requirements, disciplinary practices, and their data management needs. By providing detailed metadata about each repository, R3Data supports informed decision-making and promotes the selection of repositories that ensure compliance and long-term preservation.
When selecting a repository for research data, researchers can use a structured decision-making process starting with a decision tree to identify the most relevant repository category. This hierarchical approach begins by checking for domain or datatype-specific repositories. If none exist, researchers should consider institutional repositories, then national repositories, and finally generalist repositories as alternatives.
Identifying the appropriate category of repository is just the foundation. Selecting a specific repository within that category requires rigorous criteria. The flowchart below outlines key considerations and inquiries for evaluating a research data repository. It acknowledges that no single repository will meet all criteria perfectly and that the importance of each aspect varies based on the data's nature, regulatory requirements, and the policies of research-performing organisations.
