Development of Metadata Schemas For Collaborative Research Centers

Long-term collaborative research projects, such as Collaborative Research Centers (CRCs), generate large, heterogeneous datasets that pose significant challenges to data findability and reusability. To address this, and to comply with FAIR data principles and funding agency requirements, we have developed and implemented standardized metadata schemas for two biomedical CRCs at the University of Freiburg: CRC 1453 ("NephroGenetics") and CRC 1479 ("Oncogene-driven Immune Escape"). These CRCs focus on kidney disease genetics and tumor immune escape mechanisms, respectively, employing diverse methodologies including genomics, proteomics, imaging, and clinical studies. Our approach involved an iterative, collaborative process with researchers from both basic and clinical domains, leveraging existing ontologies and metadata standards where possible, and creating new terms when necessary. The resulting schemas, initially developed for CRC 1453 and adapted for CRC 1479, share a common structure while accommodating CRC-specific research focuses. Key modules include publication metadata, data source information (location, availability), and detailed data classification (organism, sample processing, intervention, etc.). We implemented these schemas using the Freiburg Research Data Tool (fredato), allowing researchers to annotate datasets via a user-friendly interface. For both schemas, we propose RDF knowledge graph structures and provide an ontology with the required classes and predicates. This approach enhances schema understanding, facilitates integration with other information systems, and promotes interoperability. This publication method goes beyond simple documentation, offering a blueprint for constructing semantic data models and enabling complex queries across datasets, fostering both intra- and inter-CRC collaboration and data reuse. The described method and schema are adaptable to other CRCs, contributing to long-term preservation.