FHIR as a Unifying Format for Genomic Research Data Tracking, Aggregation, and Integration

Abstract

The increasing complexity of genomic research demands standardized data sharing and integration. The Fast Healthcare Interoperability Resources (FHIR) specification has become a well-established standard for exchanging data among health data systems. While designed primarily for clinical and patient data in health care environments, it also has applicability to represent genomic research data and offers a path for aggregating and integrating extremely rich datasets that have traditionally remained siloed and disparate. To study this potential, we developed FHIR Aggregator, an integration of seven major biomedical repositories, including the Genomic Data Commons, GTEx, HTAN, and DepMap, that covered 142334 patients, 819251 specimens, 1096491 observations and 711166 documents. We explore the various ways the FHIR standard can be applied to structure genomic research data and enable new possibilities. We demonstrate how FHIR can be used, where it succeeds or falls short, and which concepts must be extended to better support large-scale clinical and genomics research projects.