This step establishes which community or domain standards, profiles, formats and serialisations will be used to represent the research object and applies and validates them in ways that support the intended uses.
Standards can make research objects easier to interpret, exchange, process and combine across people, tools, services and communities. They can define required information, structural elements, datatypes, relationships, metadata, file formats, serialisations, packaging arrangements, terminology requirements and conformance rules.
A standard is not necessarily the same as a file format. A file may use a common format without complying with a community standard, while a standard may support several formats or serialisations. A research object can require more than one standard or representation.
Standards discovery and reuse
Developing standards
Applying standards
Validating against standards
Identifier mapping
Vocabulary alignment
Model mapping
Standards discovery and reuse
Find, assess and choose existing standards, profiles, formats and serialisations that can represent the research object in ways that support the FAIRification goal and intended uses.
This capability includes identifying standards used by relevant communities, determining what each standard covers and assessing whether it can represent the selected domain model, identifiers, metadata, components and relationships.
The capability can be provided through project expertise but also through authoritative guidelines from communities, standards registries, target repositories/hosting environments, standards organisations or other infrastructure providers.
Resources
2 ELIXIR Stories items
- Plant Sciences Community Showcase Cascade mapping from FT 2–6: integration of multiple resources, data harmonisation
- Rare Disease Community Interoperability Showcase: examples from EJP RD reuse data standards: use existing RIRs where possible
3 FAIR Metroline items
- Apply common data elements Using agreed data elements turns reuse of existing standards into something practical and concrete.
- Create or reuse a semantic (meta)data model Embedding existing standards in a shared semantic model supports their reuse without redefining them.
- Creating a FAIR Implementation Profile (FIP) Recording which standards a community adopts makes reuse choices transparent and reusable.
1 FAIR Cookbook items
Developing standards
Create, profile, extend and maintain a specification where existing standards cannot adequately represent the research object or support the intended uses.
This can include creating a new standard, but it also includes less extensive approaches such as defining a community profile, adding constraints, developing an extension or proposing changes to an established standard. Standards development usually requires coordination beyond the immediate FAIRification team. Domain communities, intended users, standards maintainers, repositories, software developers and other implementers may need to participate. An important question to ask is whether the required development capability can be established through collaboration.
This capability should be considered only after relevant existing standards and profiles have been examined.
Resources
4 ELIXIR Stories items
- Plant Sciences Community Showcase Cascade mapping from FT 2–6: integration of multiple resources, data harmonisation
- Galaxy: a story about interoperability Cascade mapping from FT 4: Apply data standards: Reuse whenever possible
- Single Cell Omics Community Interoperability Showcase Cascade mapping from FT 4: Community-driven work to define standards and cross-mappings between standards through workshop programmes emerging
- Metabolomics Community & Interoperability Platform - On which subjects could they collaborate? Cascade mapping from FT 4: Developing a framework for multi-omic data analysis (transcriptomics, proteomics and metabolomics), e.g. integration of MetaboLights, UniProt, BIoModels, MetabolicAtlas, Rhea, Reactome, BIGG, and http://pickle.gr/
2 FAIR Metroline items
- Create or reuse a semantic (meta)data model Defining concepts, relations and structure supports the development of new standards where these do not yet exist.
- Apply common data elements Formalising shared data elements can be part of building a new standard.
3 FAIR Cookbook items
Applying standards
Create or transform research objects so that they conform to selected standards, profiles, formats and serialisations.
This includes applying standards when a research object is first created and applying them retrospectively to existing research objects. It may involve restructuring content, converting formats, generating metadata, packaging components, preserving identifiers and provenance, and documenting transformation decisions.
The capability can be provided through data-processing tools, export services, repository submission workflows, conversion libraries, schemas, templates, and specialists in the implementation team.
Resources
4 ELIXIR Stories items
- Plant Sciences Community Showcase Cascade mapping from FT 2–6: integration of multiple resources, data harmonisation
- Galaxy: a story about interoperability Cascade mapping from FT 4: Apply data standards: Reuse whenever possible
- Single Cell Omics Community Interoperability Showcase Cascade mapping from FT 4: Community-driven work to define standards and cross-mappings between standards through workshop programmes emerging
- Metabolomics Community & Interoperability Platform - On which subjects could they collaborate? Cascade mapping from FT 4: Developing a framework for multi-omic data analysis (transcriptomics, proteomics and metabolomics), e.g. integration of MetaboLights, UniProt, BIoModels, MetabolicAtlas, Rhea, Reactome, BIGG, and http://pickle.gr/
4 FAIR Metroline items
- Apply common data elements Consistent use of agreed data elements is one of the main ways data standards are applied in practice.
- Apply (meta)data model Implementing the agreed model in data and metadata makes data standards actionable.
- Transform and expose FAIR (meta)data Turning data and metadata into a more structured FAIR form helps put standards into practice.
- Design eCRF (data collection) Building standards directly into eCRF design connects application of standards to data collection.
3 FAIR Cookbook items
Validating against standards
Assess whether a research object representation conforms to the requirements of a specified standard, version, profile or extension.
This can include syntactic, structural, schema-based, semantic, rules-based, completeness and cross-component validation. Different validation mechanisms may be required for different parts of a compound research object.
Validation establishes conformance with defined requirements. It does not by itself establish that the research object is FAIR, scientifically correct or suitable for every intended use.
The capability can be provided through official validators, schemas, repository submission checks, testing frameworks, quality-control workflows, community services or manual expert review.
Resources
4 ELIXIR Stories items
- Plant Sciences Community Showcase Cascade mapping from FT 2–6: integration of multiple resources, data harmonisation
- Galaxy: a story about interoperability Cascade mapping from FT 4: Apply data standards: Reuse whenever possible
- Single Cell Omics Community Interoperability Showcase Cascade mapping from FT 4: Community-driven work to define standards and cross-mappings between standards through workshop programmes emerging
- Metabolomics Community & Interoperability Platform - On which subjects could they collaborate? Cascade mapping from FT 4: Developing a framework for multi-omic data analysis (transcriptomics, proteomics and metabolomics), e.g. integration of MetaboLights, UniProt, BIoModels, MetabolicAtlas, Rhea, Reactome, BIGG, and http://pickle.gr/
3 FAIR Metroline items
- Pre-FAIR assessment An early assessment helps reveal whether data already aligns with relevant standards and where gaps remain.
- Assess FAIRness Checking whether data and metadata meet FAIR requirements complements validation against standards.
- Design eCRF (data collection) Structured forms can support validation by enforcing standard compliant input at the moment of collection.