Three Levels of Connectivity

Level 1: Resource Discovery

Description

Contributes to

Technology Requirements

Access condition options

Example Queries

At this level, the provider commits to openly publish online some standardised metadata about the offered resource, and hence make this available to the VP via the VP Index.

Resource discoverability via open metadata

FAIR Data Point specification, EJP RD metadata schema

Open Access

What are the URLs of Catalogs that allow deeper queries? What are the available biobanks? What are the available patient registries?

Prerequisite: Level 1 connectivity

Description

Contributes to

Technology Requirements

Access condition options

Example Queries

At this level, a resource is identified based on remote queries regarding its characteristics and content, responding with yes/no or approximate record count information. The questions are answered against summary metadata and safe content of each catalogue/resource.

Resource discoverability via controlled querying of catalogue/resource summary info and/or safe resource record data

EJP RD Beacon v2 API implementation

Open or authenticated user access, as per the preference of each resource

‎ ‎

Level 2 – Querying at the catalogue/resource level

Involves answering queries based on the summary level metadata of RD resources

EJP RD queryable metadata

Open user access

Is the catalogue associated with the Marfan syndrome [ordo:Orphanet_558]?

Level 2 - Querying at safe-record level

Entails answering queries based on individual records of resource data.

EJP RD queryable data (obfuscated record data)

Open or authenticated user access, as per the preference of each resource

Find resources based upon how many patients have Autosomal recessive polycystic kidney disease [ORPHA:731] and had symptom onset before 8 years old?

Prerequisite: Level 1 connectivity

Description

Contributes to

Technology Requirements

Access condition options

Example Queries

At this level, the provider commits to support interrogation and analysis on its resource’s rich content.

Data reuse and analysis

SPARQL, FAIR Data Train, Data available according to the Clinical And Registry Entries Semantic Model (CARE-SM)

Open access, Authentication, Authorization

Training a prediction model on distributed data.