Show summary
INTRODUCTION: Open sharing of scientific results is widely recognized as crucial for advancing neuroscience research, yet the quality and the structure of shared outputs remain inconsistent. While researchers are increasingly willing to share their data, many lack the necessary training, expertise, and support for proper data management practices [1,2]. Additionally, the variety of standards and formats can complicate the adoption of appropriate data and metadata structures, often making data from different sources difficult to compare and combine. To address these challenges, the EBRAINS Research Infrastructure has developed a standardized data curation workflow built on a structured metadata framework, with recent updates aimed at improving usability, accessibility and efficiency for researchers. The workflow ensures that outputs are well-structured, discoverable, interpretable, and reusable, ultimately facilitating better integration across diverse platforms while adhering to FAIR principles.
METHODS: Continuously shaped by researcher needs, the EBRAINS Data Services employ a structured, user-friendly curation workflow to facilitate FAIR-aligned data sharing across a broad range of neuroscience domains. The workflow includes both a human-readable data descriptor to support dataset interpretation, and machine actionable metadata annotation based on the openMINDS framework (RRID: SCR_023173), ensuring interoperability and reusability. The workflow design has been informed by iterative user feedback. To simplify the metadata integration, an intuitive metadata wizard (https://metadata-wizard.apps.ebrains.eu/) has been developed. This tool assists researchers in submitting metadata that can be readily converted to openMINDS. The improvements to the workflow have been iteratively refined through user feedback and validated across multiple neuroscience subdomains, demonstrating its adaptability and effectiveness.
RESULTS: Here we present recent improvements to the EBRAINS curation workflow designed to enhance metadata quality, usability, and data accessibility. Key updates include enhancements to the Metadata Wizard include dynamic term retrieval from the EBRAINS Knowledge Graph and new features like secure authentication, collaborative editing, a guided data descriptor questionnaire, and enhanced support for metadata collection. To further support researchers, the curation workflow improves the discoverability and interpretability of sensitive data by sharing data descriptors and anonymous metadata. This approach makes sensitive data accessible via sensitive data infrastructures following an application procedure, ensuring compliance with ethical guidelines and regulations. These improvements to metadata quality and accessibility lay the foundation for integrating EBRAINS datasets with external research infrastructures. Building on this foundation, EBRAINS datasets are now connected to the broader ecosystem of neuroscience research outputs in the OpenAIRE Graph through the SciLake Neuroscience pilot which leverages graph based methods to structure and contextualize scientific information and support analysis of dataset usage, emerging trends, and research visibility.
DISCUSSION: These recent updates strengthen the EBRAINS curation workflow by improving metadata integration, streamlining submission processes, and enabling broader support for diverse and sensitive data types. The improved curation workflow reduces the complexity of data sharing for researchers while ensuring compliance with ethical and technical standards. Integration with SciLake and OpenAIRE further extends the reach of EBRAINS datasets, providing insights into usage and impact that support both service refinement and increased research visibility. As a result, EBRAINS is increasingly positioned as a driving force in open and collaborative neuroscience, with continuous improvements guided by the evolving needs of the research community.
