Community Cloud Integration

Background

The Community Cloud project and the Genomic Data and Access Management (GDAM) API aim to facilitate data sharing and use of genomic data files through:

  • Capture of Genomic Study metadata, including analysis performed and devices used (using the GenomicStudy and GenomicStudyAnalysis backport profiles developed by the HL7 Clinical Genomics WG)
  • Capture of links to Genomic Data files, using the DocumentReference resource and potentially the GA4GH DRS protocol to abstract the underlying data store's structure. GA4GH htsget can also be used to query specific locations within genomic data files, and GA4GH BEACON could also potentially be supported to support querying across datasets, e.g. for cohort identification.

FHIR API

Community Cloud

The Community Cloud project has identified a minimal set of metadata to be captured alongside genomic data files, recreated in the table below. This has been mapped to a set of FHIR resources.

Data Element Example FHIR Mapping
Clinical test identifier (e.g. R-code / M-code) R134 Procedure(GenomicStudy).reasonCode.coding.code
Clinical indication name (e.g. Thanatophoric dysplasia) Familial Hypercholesterolaemia Procedure(GenomicStudy).reasonCode.coding.display
Disease domain (rare disease / solid cancer / haemonc) Rare Disease Procedure(GenomicStudy).note (pending removal as can be inferred from test identifier)
Sample identifier (stable within submitting organisation) 140103896-25301R0078 Specimen.identifier.value
VCF identifier 140103896-25301R0078-25NGCEN64-9527-F-99347387_markdup_recalibrated_Haplotyper.vcf.gz DocumentReference.content.attachment.title
Molecule type (DNA / RNA) DNA Specimen.type.coding (using appropriate SNOMED CT code from Genomic-Specimen-FinalSample
Submitting organisation (GLH and specific team) East GLH Cambridge DocumentReference.author (using GLH ODS Code TBC Trust ODS code may be required in addition to the GLH ODS code)
Sequencing platform (i.e. Illumina or Aviti) Illumina Device.manufacturer
Sequencing instrument model (e.g. Novaseq 6000) Novaseq 6000 Device.deviceName.name
Sequencing assay name (e.g. Twist Exome v3) CEN Device.property.valueCode.text (where Device.property.type.text="Sequencing assay name") TBC whether this falls under the same device as the instrument model
Bioinformatics pipeline name Dias Device.deviceName.name (separate device for the Bioinformatics pipeline)
Bioinformatics pipeline version eggd_conductor_dias_CEN_config_v2.5.2 Device.version.value
Bioinformatics variant caller name Sentieon Haplotyper Device.deviceName.name (separate device for the Bioinformatics variant caller)
Bioinformatics variant caller version 202308.03 Device.version.value
Bioinformatics variant type represented (small / CNV / SV / fusion) Small Procedure(GenomicStudy).code.text TBC May be coded if aligning to the GenomicStudy Type ValueSet or moved to Procedure(GenomicStudyAnalysis) if aligning to the GenomicStudyAnalysis MethodType or ChangeType ValueSets
Sequencing run date 2025-11-25T17:49:00Z Procedure(GenomicStudy).performedDateTime
S3 URL https://uk01.l.antigena.com/l/MA77... DocumentReference.content.attachment.url

A set of FHIR Examples have been created to facilitate visualising the resources. Please note, that until integrated with Genomic Order Management, the resources produced for the CommunityCloud may not contain the recommended set of elements as identified elsewhere in the IG e.g. Specimen collection details. Alignment and ability to capture/import Genomic Order Management related data will be investigated in future phases.

Resources follow the following referencing structure:

Genomic Data and Access Management (GDAM)

The GDAM API uses a subset of information to generate the reference to the Genomic Data File. In terms of FHIR, only the DocumentReference resource is used. Guidance on how to populate the resource is provided on the DocumentReference profile page.

The sequence diagram below shows the expected sequence of events for both saving a reference to a Genomic Data File and querying the file via the GDAM API:

GDAMSequence