2 February 2022
Healthcare systems capture vital clinical information, such as a patient’s disease symptoms, observable characteristics, and demographic data. Improving patient care and clinical outcomes rely on the ability to standardize and share this information, along with its corresponding genomic data.
First approved in 2019, the GA4GH Phenopackets standard offers a human and machine-readable way to structure phenotypic data about a patient or individual. This information can then be shared across clinical and research environments or used for computational analyses.
To expand the utility of the standard, the Phenopackets development team has now released Phenopackets v2.0. Originally developed for the rare disease use case, version two enables better representation of cancer and common disease as well. As software becomes an increasingly powerful tool in genomic medicine, the team aims to unite genomic and phenotypic data for analysis across research and healthcare—a concept further discussed in the team’s preprint, “The GA4GH Phenopacket schema: A computable representation of clinical data for precision medicine.”
“Phenopackets is a community-driven standard,” said Melissa Haendel, co-lead of the Clinical and Phenotypic Data Capture Work Stream, Chief Research Informatics Officer at the University of Colorado Anschutz Medical Campus, and Principal Investigator on the NHGRI Phenomics First program funding Phenopackets. “We hope Phenopackets can play a role in bridging the research and clinical spaces and deliver on the promise of personalized healthcare.”
Phenopackets v2.0 includes the addition of three new elements, allowing the capture and sharing of a much more complete medical picture:
“The new Measurements, Medical Actions, and Time Element fields enable capturing patient information around common disease and cancer in a much more precise and expressive manner than with v1.0,” said Jules Jacobsen, co-lead of the Phenopackets development team and senior software developer at the Queen Mary University of London. “These changes are especially beneficial for modeling time-courses, for example of disease progression and response to therapeutic regimens. They also enable greater interoperability with clinical data models such as the Observational Medical Outcomes Partnership (OMOP) Common Data Model.”
Núria Queralt Rosinach, a contributor on the GA4GH Clinical and Phenotypic Data Capture Work Stream, member of the European Joint Programme on Rare Diseases (EJP RD) Driver Project, and researcher at Leiden University Medical Center, stated, “Quantitative phenotypes such as clinical measurements are of paramount importance to monitor the health status of patients. For instance, following cytokine levels over time to characterize the immune-response of patients during the disease course can aid clinicians on prognosis and intervention decision-making in cases such as COVID-19.”
Through a COVID-19 BioHackathon and further discussions at Phenopackets meetings, Rosinach’s team proposed the new Measurements element—an example of the value of open, community-driven, and collaborative science within the GA4GH community.
To cater to better cancer diagnostics, the Phenopackets team collaborated with ICGC-ARGO, an international organization leveraging genomics to accelerate cancer research, and mCODE, an initiative aiming to create a common cancer data model, to add cancer-relevant fields, ensuring utility of the standard to the cancer community.
Another major goal of version 2.0 is aligning with other standards. To that end, Phenopackets v2.0 integrates the Variation Representation Specification (VRS) and VRSATILE—related technical products developed by the GA4GH Genomic Knowledge Standards Work Stream that aim to standardize the exchange of genetic variation data. This integration serves to reorganize Phenopackets’ previous interpretation elements to better support semantics for rare disease and cancer variant interpretation.
“By collaborating with the VRS team, we have an integrated and sophisticated model for the two most important domains in diagnostic genomics—genotype and phenotype—now all using GA4GH standards,” said Peter Robinson, co-lead of the Phenopackets development team and professor of computational biology at the Jackson Laboratories.
“Integrating VRS and Phenopackets is a key advancement that unites two GA4GH standards designed under different messaging frameworks,” continued Alex Wagner, co-lead of VRS and Assistant Professor at the Institute for Genomic Medicine at Nationwide Children’s Hospital. “We should always strive to integrate GA4GH standards where appropriate, as this greatly improves interoperability between downstream tools and processes that leverage these standards.”
To further interoperability efforts with other standards in the community, the team is developing a Phenopackets-FHIR Implementation Guide to ensure compatibility with Fast Healthcare Interoperability Resource (FHIR) HL7—a widely-adopted standard for exchanging electronic health records.
The team has also brought Phenopackets to the International Organization for Standardization (ISO) Technical Sub-Committee for Genomics Informatics (ISO TC/21/SC1). This global collaboration aims to support Phenopackets by making standardized phenotypic information more broadly available and supporting even more use cases around the world.