The Monarch Initiative at the 12th GA4GH Plenary Meeting

Photo: Mónica Muñoz Torres, Nicole Vasilevsky, and Tudor Groza at the GA4GH Plenary Meeting on September 16, 2024

The Global Alliance for Genomics and Health (GA4GH) aims to accelerate progress in genomic science and human health by developing standards and framing policy for responsible genomic and health-related data sharing. The Monarch Initiative is a Driver Project of GA4GH. In this capacity, we leverage our expertise in translational and integrative sciences at the intersection of GA4GH’s central themes: Basic Biology, Cancer, Complex Traits, and Rare Diseases research.

The 12th GA4GH Plenary Meeting was co-hosted with Australian Genomics and held in Melbourne, Australia, on September 16–20, 2024. As part of this meeting, GA4GH held its biannual Connect event, a collection of working sessions that allow contributors to advance the GA4GH Road Map and gather feedback on development and needs for standards and policies. The organizers opened the meeting by inviting all participants to accompany them in acknowledging the Traditional Custodians of country throughout Australia and their connections to land, sea, and community. Participants paid their respect to their elders, past and present, and extended that respect to all Aboriginal and Torres Strait Islander peoples today.

The agendas for the GA4GH Plenary Meeting and the Connect event, along with links to presentations and recordings, can be found on the conference website. Below are some topics discussed at the meeting that illustrate how we at the Monarch Initiative contribute to advancing GA4GH’s mission.

Clinical and Phenotypic Data Capture Work Stream

GA4GH’s Clinical & Phenotypic Data Capture Work Stream (Clin/Pheno) accelerates the use of genomics in medicine by standardizing how we capture, describe, and exchange clinical data, focusing on observable traits (phenotypes) and family health history information. Since 2023, Monarch’s Dr. Mónica Muñoz Torres has co-led this Work Stream with Dr. Mélanie Courtot from OICR. To assist in achieving this Work Stream’s goals, the Monarch Initiative contributes to developing the Phenopackets schema and two family health history standards, the Pedigree standard and the Family History Relations (KIN) Ontology.

At the Connect sessions, Dr. Muñoz Torres led a discussion focusing on current obstacles to implementing GA4GH Clin/Pheno standards, explored the available solutions, shared insights and experiences from use cases, introduced new implementers to the Work Stream’s products, and initiated the process of collaboratively developing a strategic roadmap for enhancing the implementation of Clin/Pheno standards.

Monarch team members play critical roles in several of the standards and software products developed or leveraged by the Clin/Pheno Work Stream:

Phenopackets. The Phenopacket schema, developed by GA4GH, is an ISO-approved, human, and machine-readable format for structuring clinical and phenotypic data about a patient or individual. This project involves participation from Monarch’s Drs. Jules Jacobsen, Peter Robinson, Mónica Muñoz Torres, Shahim Essaid, and Melissa Haendel.

Phenotypic and clinical data representation using ontologies: cross references, mappings, and beyond. Monarch team members Drs. Peter Robinson, Tudor Groza, and Mónica Muñoz Torres offered examples of available tools to address the significant challenge of mapping clinical data about phenotypes and diseases collected in existing cohorts, often encoded in non-standardized formats.

• Dr. Robinson discussed software tools developed to help export tabular data into the Phenopacket format using Mondo and HPO terms, such as phenopacket tools.
• Dr. Groza described the effective text mining software he has developed for extracting HPO terms from EHR texts — FastHPOCR: pragmatic, fast, and accurate concept recognition using the human phenotype ontology.
• We introduced Monarch’s ongoing work, led by Dr. Melissa Haendel and Dr. Nicolas Matentzoglu, on creating a comprehensive mapping between SNOMED-CT and HPO to address overlapping clinical findings and meet the growing user demands for both resources.

Representation of family relationships. Drs. Mónica Muñoz Torres, Chris Mungall, Jim Balhoff, and Matthew Brush currently contribute to developing the Pedigree standard and the Family History Relations (KIN) Ontology.

Members of the Monarch Initiative also highlighted the development of standards and tools to improve the identification and diagnosis of rare diseases (see section below).

Variant Annotation Specification

The Variant Annotation (VA) subgroup of the Genomic Knowledge Standards work stream is tasked with developing standards for the representation and exchange of information about genetic variation. It aspires to support explicit and computable representation of clinically relevant knowledge, such as variant pathogenicity or therapeutic response classifications, along with evidence and provenance that enable assessment of the relevance and utility of such knowledge.

The Monarch Initiative is a key contributor to this work as an active Driver Project and through the participation of Matthew Brush as co-lead of the VA effort. At the Connect meeting, the VA team announced a draft release of the Variant Annotation Specification (VA-Spec) — which included an initial set of standard schema (‘profiles’) for representing seven kinds of variant knowledge, along with a modeling framework that allows for implementation-based, community-driven development. Each schema was developed through a partnership between VA leads and one or more driver projects, which actively implemented it to support real datasets they were producing. This initial test of the VA development framework will help elicit feedback that informs the expansion of initial models, the development of standards for new types of variant knowledge, and the formalization of the modeling framework.

GA4GH welcomes all members of the research or clinical community who wish to follow or participate in the ongoing development, application, and testing of the VA specification. The VA team holds monthly recorded community calls, hosts an active set of Slack channels, and has a responsive group of leads keen to engage new contributors. Reach out to Matthew Brush for more information.

Rare Disease Community of Interest

The GA4GH Rare Disease Community of Interest (RD COI) was established in 2022 with Drs. Tudor Groza and Nicole Vasilevsky assuming the roles of co-chairs in 2023. The GA4GH RD COI serves as a central hub for rare disease experts and global initiatives, fostering collaboration to address real-world challenges in rare disease through the application of GA4GH standards.

By connecting rare disease genomics initiatives focused on data infrastructure, management, analysis, and sharing, this community is uniquely positioned to bridge the gap between clinical and technical experts. GA4GH has developed a significant number of products and standards to support the operationalization of genomic health implementations and advance the broad field of genomic medicine. While these standards offer wide-ranging, domain-agnostic applications, their diverse scope and overlapping functionalities can pose challenges when creating tailored, cohesive solutions for specific domains, such as rare disease.

One of the primary goals of the RD COI is to curate a GA4GH portfolio of products and provide guidance on their recommended usage within the rare disease domain. At GA4GH Connect this past September, the RD Community hosted a Work Stream “tour” to highlight the substantial work and available products applicable to rare diseases. This included presentations by:

1. Dr. Robert Freimuth on the Genomic Knowledge Standards (GKS) Work stream
2. Drs. Dianne Nicol & Diya Uberoi on the Regulatory & Ethics Work Stream (REWS)
3. Drs. Mónica Muñoz Torres & Mélanie Courtot on the Clinical & Phenotypic Data Capture Work Stream (Clin/Pheno)
4. Dr. Vasiliki Rahimzadeh on the Data Use & Researcher Identity (DURI) Work Stream
5. Dr. Jaime Delgado on the Data Security Work Stream

Several GA4GH standards have been identified as relevant to the Rare Disease domain. These include:

1. The Genomic Knowledge Standards (GKS) work stream has produced three standards to facilitate the representation and effect prediction of variants: the Variant Representation Specification VRS), the Categorical Variation Representation Specification (Cat-VRS), and the Variant Annotation (VA) specification.
2. The Regulatory and Ethics Work Stream (REWS) has produced model consent clauses for Rare Disease Research (see also this paper) and issued a Diversity in Datasets Policy.
3. Clin/Pheno has produced the Phenopackets standard.
4. The Data Use and Researcher Identity (DURI) Work Stream has created the Data Use Ontology (DUO) and the GA4GH Passports.
5. Finally, the Data Security Work Stream has produced the Authorization and Authentication Infrastructure (AAI).

In particular, standards and tools spearheaded by the Monarch Initiative in the context of the Clin/Pheno work stream — such as Phenopackets, mentioned above — as well as others being developed by the Monarch team — such as the Human Phenotype Ontology and Mondo — will play a critical role in enabling data sharing and harmonization in the field. Next, the RD COI will convene leaders from each Work Stream to build the GA4GH Rare Disease Portfolio, a catalog of existing standards relevant to the RD community.

As we look toward the future of the RD COI, we created a long-term roadmap focusing on three key areas: 1) strengthening community engagement, 2) expanding GA4GH standards implementation in the rare disease field, and 3) fostering collaborative research and development. We invite all interested individuals to join our Rare Disease community or contribute to the portfolio. For additional information, please contact Nicole Vasilevsky (nicole.a.vasilevsky@gmail.com) or Tudor Groza (tudor.groza@gmail.com).