GA4GH QC-WGS Sep'25 Meeting

[justinjj24]

Agenda

Decision

1. Metric Definition Attributes
   o Extend the metric definition template with attributes:

   • Data type (e.g., WGS Short-read/Long-read, RNA-Seq, single-cell)
   • Assembly/reference sequence collection (aligned with GA4GH Refget terminology rather than just "assembly")
   • Version (of the metric definition and of the reference implementation/tool)
   • External references and access IDs where applicable

2. Schema Flexibility
   o JSON schema will allow for attributes that adapt across different data types and technologies.
   o Include placeholders for implementation details while distinguishing them from version-controlled definitions.

3. Terminology Alignment
   o Use the term “reference genome sequence collection” (per GA4GH standards) instead of "assembly."
   o Use “version” consistently (instead of "tool") to avoid ambiguity.

4. Backward Compatibility
   o Maintain compatibility with hg19/GRCh37 to meet PRC and adapter requirements.
   o Explicitly track which genome builds a given metric definition supports.

5. Focus of Metrics
   o Prioritize broad quality assessment metrics (coverage, duplication, contamination) before complex site-specific metrics.
   o Document site-specific extensions as optional, not required for baseline adoption.

6. Public Outreach and Roadmap
   o Gather feedback from clinical genomics laboratories community to refine applicability.
   o Explore support for long read sequencing and somatic QC metrics in future versions.
   o Provide real-world examples of QC metric implementation (~PRECISE data)
   o Develop validation tools to check compliance against JSON schema.

Rationale

• Interoperability: Defining attributes like data type, reference collection, and version ensures metrics can be consistently interpreted across datasets.
• Clarity: Separating implementation details from versioned definitions reduces confusion when pipelines evolve.
• Alignment: Using GA4GH-compliant terminology (e.g., "reference genome sequence collection") avoids schema drift.
• Adoption: Clinical genomics groups require backward compatibility (hg19) and simple, clear metrics for implementation.
• Future-proofing: The schema design anticipates additional sequencing technologies without requiring disruptive changes
• Metrics: discusses the challenges of implementing site-specific metrics and suggests focusing on broader quality assessments.