The FAIR Cookbook - the essential resource for and by FAIR doers

This paper addresses how to operationalize the FAIR Principles (Findable, Accessible, Interoperable, Reusable) in practice within the Life Sciences by introducing and describing the FAIR Cookbook. While FAIR has been widely endorsed by funders, publishers, infrastructures, and industry, organizations struggle to translate these high-level, aspirational principles into concrete, context-appropriate actions, and to quantify the costs and benefits of FAIR implementations. The authors identify two core challenges: plotting practical, domain-appropriate FAIRification pathways in the absence of a single standard or technology, and evaluating the return on investment to motivate organizational change. The FAIR Cookbook is presented as a pragmatic, community-built solution that provides hands-on recipes, examples, and guidance to bridge the gap between principles and implementation, support capability building, and drive culture change toward FAIR data practices.

The FAIR Cookbook was developed and maintained using open, off-the-shelf technologies and community workflows to minimize cost and maximize sustainability. The site is built with Jupyter Book, with version control, hosting, continuous integration, and automated builds via GitHub and custom GitHub Actions. Content is authored primarily in Markdown using HackMD for collaborative editing and review, with additional contribution routes through shared documents or direct GitHub pull requests. Executable components are provided as Jupyter Notebooks integrated into recipes and runnable via Binder or Google Colaboratory. Visual elements include standardized summary cards, Font Awesome icons, and Mermaid-generated diagrams; images are managed as MMD sources rendered to PNG. Recipes are designed to be citable and credit authors: persistent identifiers are minted via w3id.org, author identification uses ORCID, and roles are captured with the CRediT taxonomy. The resource itself is registered in identifiers.org with a dedicated namespace. Recipes are annotated with FAIR Dataset Maturity (FAIR-DSM) model indicators to signal the maturity aspects and levels attainable through each recipe. Findability and discoverability are enhanced by SEO measures (sitemap via sphinx-sitemap), JSON-LD markup using Bioschemas TrainingMaterial and schema.org HowTo, and a custom search wizard enabling filtering by recipe metadata (type, audience, reading time, maturity level, code availability). The resource adheres to FAIR itself: persistent identifiers per recipe, standardized metadata, public HTTPS access, JSON-LD interoperability, cross-links to registries such as FAIRsharing, and CC BY 4.0 licensing. Governance includes an Editorial Board steering content creation, review, and quality control; content selection combined top-down planning of a table of contents with bottom-up triage of needs from IMI/IHI projects and industry partners. Contributions are coordinated via regular calls, book-dash events, GitHub milestones, and communication channels (Slack, email, GitHub issues). Review mirrors scholarly peer review with assigned reviewers assessing coverage, syntax, code presence, and reproducibility; both authors and reviewers are credited. Operations include continuous integration for automated releases, Zenodo integration to mint DOIs per release, and citation metadata via the Citation File Format. A Docker image enables local deployment and testing, facilitating organizational customization and on-premises instances.

Content and coverage: As of February 2023, the FAIR Cookbook contains over 82 production-grade recipes. Approximately one fifth of recipes include associated executable code runnable in the cloud. Recipes span general FAIR principles, software infrastructure (e.g., ontology browsers, data catalogs), FAIR assessment tools, and applied exemplars across molecular, pre-clinical, and clinical domains. Contributors and community: Nearly 100 contributors from more than 40 academic and industrial organizations have authored content, coordinated through ELIXIR Nodes and project partners. Usage: Between January 2022 and January 2023, the site had 13,889 users and 50,418 page views (Google Analytics). Utility and impact: - Education: Used in the FAIRplus Fellowship Programme to train 20 fellows who both consumed and produced recipes, validating learning outcomes despite varying technical backgrounds. - Industry practice: Janssen used recipes and maturity indicators to evaluate data lake design and anticipated improvements in metadata richness, standardization, interlinking, and unified access, aiding ROI discussions and investment justification. Boehringer Ingelheim demonstrated data integration using open ontologies and application ontology construction via open tools and OBO resources. AstraZeneca co-developed a recipe on digital rights (FCB035) aligning DCAT/RDF-based metadata with enterprise clinical data access policy. Adoptions and endorsements: - Recognized and recommended by EC/IMI/IHI guidelines for open access to publications and research data, and cited in an EC report on FAIR data for coordinated COVID-19 responses; included in Horizon Europe Work Programme 2023–2024 for Health. - Became an ELIXIR service (UK and Luxembourg Nodes in 2021; Switzerland in 2022; Spain in 2023) and is listed among ELIXIR flagship resources; embedded in ELIXIR Scientific Programme 2024–2028. - Integrated and cross-linked with RDMkit, Pistoia Alliance’s FAIR Toolkit, and FAIRsharing; ongoing work to tag recipes with Terms4FAIRskills competencies. Technical achievements: Implemented persistent identifiers (w3id), author credit (ORCID, CRediT), maturity indicators (FAIR-DSM), SEO and structured data (Bioschemas, schema.org), a custom search wizard, automated releases with Zenodo DOIs, and containerized deployment.

The FAIR Cookbook addresses a critical gap between aspirational FAIR Principles and actionable, domain-relevant implementation guidance. By providing citable, peer-reviewed, and executable recipes authored by practitioners across academia and industry, it enables users to build coherent FAIRification pathways aligned with maturity targets and organizational goals. The resource serves multiple roles: a practical manual for day-to-day FAIR data operations, a curricular component for training programs, and a catalyst for organizational culture change towards FAIR stewardship. Its integration with the FAIR-DSM model offers a means to plan and assess progress along the FAIR continuum, supporting cost-benefit considerations and internal advocacy (e.g., ROI discussions). The Cookbook’s positioning within the ELIXIR ecosystem and its endorsements by major infrastructures and funders reinforce its credibility and sustainability. Collaborative ties with initiatives like RDMkit, the Pistoia Alliance’s FAIR Toolkit, FAIRsharing, and the NIH Office of Data Science Strategy extend its reach and foster convergence across complementary resources. The participatory model—through book dashes, editorial oversight, and open infrastructure—unlocks collective expertise and harmonizes practices (e.g., FAIRification of observational studies), while inviting ongoing expansion into specialized domains and enterprise contexts.

The FAIR Cookbook delivers specialized, hands-on guidance that operationalizes FAIR in Life Sciences, bridging the gap between high-level principles and real-world implementation. It has rapidly become a flagship, community-driven, and citable resource with measurable educational, practical, and strategic value across public and private sectors. Sustained by lightweight, open infrastructure and embedded in the ELIXIR framework, it benefits from endorsements by funders and collaborations with global initiatives. Future directions include strengthening governance (e.g., Domain Boards), supporting networks of public and organization-specific instances, deepening coverage in specialized domains (e.g., knowledge graphs, semantic web), enhancing maturity guidance and goal-oriented navigation, and systematically collecting user feedback to improve the user journey. The overarching aim is to make FAIR less intimidating and more achievable through credited contributions, community convergence, and recognition of diverse FAIRification paths.

Key challenges include: (1) the inherent aspirational nature of FAIR leading to variability in implementation routes and the need for continuous updates; (2) uneven technical background requirements—some recipes have a steep learning curve; (3) sustainability of content—authoring, reviewing, and maintaining up-to-date recipes requires committed expert time; (4) feature development—implementing new functionalities beyond the lightweight infrastructure may require dedicated funding; (5) governance and scaling—establishing structures that meet the needs of a growing, diverse contributor community; (6) coverage gaps—recipes do not yet span all enterprise-specific processes or domains, and private-sector constraints may necessitate internal instances for proprietary topics; (7) cost-benefit assessment remains context-dependent and can be difficult to quantify uniformly.