ClinVar-GKM Primer¶

A ~8-minute (15-slide) introduction to ClinVar-GKM for a technical audience that knows ClinVar but is new to VA-Spec and the GKM model. It covers how ClinVar's complete data is organized today (shown with the real VCV XML schema), how the newest GKM standards (VRS 2.1.1 · Cat-VRS 1.1.1 · VA-Spec 1.1.0, on GKM-Core 1.3.0) reorganize it for consistent, accurate access, the bundle shape and class inventory, a worked example, the community profiles and ClinVar-specific implementer profiles in play, downloads, the transparency of how the data is produced, and who is adopting it for production.

Open the presentation (fullscreen) ↗

The deck is interactive — use the arrow keys to navigate and hover any info chip for a definition. The preview below runs the same deck inline; open it fullscreen (link above) for presenting.

Keyboard shortcuts¶

Agenda¶

1. The starting point — ClinVar's complete truth is the weekly XML
2. Anatomy of the XML — what "just extract the SCVs" really takes (real VCV schema)
3. What it is — every ClinVar release rebuilt as one GKM bundle
4. The GKM stack — VRS 2.1.1 · Cat-VRS 1.1.1 · VA-Spec 1.1.0 · GKM-Core 1.3.0 (all newly released)
5. ClinVar → statements — SCV / VCV / RCV as VA-Spec Statements
6. The GKM Bundle — compaction via ids + #/section/id pointers (like foreign keys); schema-driven Starter Kit / Toolkit
7. Class inventory — the GKM classes in play (MappableConcept, ConceptSet, CategoricalVariant, Allele, CNV Count/Change, Statement, Proposition, EvidenceLine)
8. Worked example — clinvar:10 (HFE p.His63Asp), statement → proposition → variant → VRS
9. Profiles & recipes — Cat-VRS recipes and VA-Spec community profiles
10. Implementer profiles — custom propositions + ClinVar-specific extensions
11. Downloads — monthly full + weekly deltas, JSON & Parquet
12. Transparency — how every value is produced and verified
13. Momentum — groups adopting it for production
14. Get started — docs, libraries, and the feedback loop