DRS and FHIR: integrating different standards to power complex scientific research in the INCLUDE DCC Project

The INCLUDE Data Hub provides researchers with comprehensive clinical, imaging, and genomic data for persons with Down syndrome (trisomy 21). The data have been standardized using FHIR for the clinical data, and DRS for the genomics data. These standards are used by the INCLUDE Portal and CAVATICA to power the discoverability of the data, making them Findable, Accessible, Interoperable and Reusable (FAIR). The Include Portal provides a user-friendly interface to discover the data; the user creates the cohort of interest and then sends the data to CAVATICA for analysis. The operation is carried out using the DRS bulk import API deployed on CAVATICA, which provides an easy way to import a list of DRS_URI and also associated metadata. A fhir_document_reference is associated with each DRS URI; it contains the link to the NCPI DocumentReference FHIR Resource. This structure provides a two way connection between the source of the clinical data, and the source of the genomics, allowing: (i) avoid data duplication, keeping one source of truth for the data of competence in each system, (ii) future-proofing the ability to obtain and search complex metadata directly on the FHIR server using the fhir_document_reference as the connection on each DRS URI, empowering researchers and clinicians to investigate any of the metadata, removing the constraints attached on importing just a subselection of metadata or agreeing on that subset. As an example we demonstrate how a user can obtain complex traits, like presence of trisomy 21, directly from the FHIR Server using the MONDO ontology via the FHIR API, the case id and the sample id, and assign them to the DRS URI on CAVATICA. This is the starting point to perform a bulk RNA DESeq analysis comparing patients using the trisomy 21 state as differentiator to obtain which genes are up or down-regulated between the two conditions.