Skip to content

Roadmap

ClinVar-GKM's direction is shaped in the open. The table below is the current backlog of ideas under consideration, ordered by community upvotes. Each idea has a matching GitHub Discussion — open it and use the Upvote button (the ▲ in the bottom-left corner of the discussion's opening post) to signal you want it prioritized, and comment with your use case. Maintainers use the upvote tally to decide the order in which features and improvements are taken on.

Nothing here is a commitment or a delivery date — it's a prioritization board. Items move through Proposed → Spec'd → In progress → Shipped as they gain support and get built.

Current backlog

The table is generated automatically from the GitHub Discussions labeled roadmap authorized, ordered by upvote count. To read the full proposal (and any linked spec) for an item, open its discussion.

# Idea Status Upvotes Discussion
1 VCV/RCV clinical-significance & star-level aggregate extensions — Add three additive extensions to the aggregate (VCV & RCV) statements so consumers can query review quality and concordance/conflict directly: reviewStarRating… Spec'd 1 #99
2 Starter-kit documentation page — A docs page showing how to consume the published bundle + Parquet with the GA4GH Python libraries (vrs-python, cat-vrs-python, va-spec-python) — load-from-bund… Proposed 1 #100
3 Upstream clinvar-ingest pipeline documentation — Document the upstream clinvar-ingest project: how it parses each released ClinVar XML dataset into JSON + relational BigQuery tables (the clinvar_ingest datase… Proposed 1 #101
4 Expert Panel / Practice Guideline VCV classification fix — Fix EP and PG VCV statements so that when multiple contributing SCVs exist they produce distinct, correct aggregate classifications rather than collapsing. Proposed 1 #102
5 Release browser reorganization + weekly Parquet + archiving — Reorganize the Downloads "Browse All Releases" widget into a clear 2×2 — Bundles (Monthly/Full, Weekly/Deltas) and Parquet (Monthly/Full, Weekly/Deltas) — and … Spec'd 1 #112
6 Per-release exception log + processing-policy reference — Produce a per-release exception/anomaly log (vrsify + proc issues) and a comprehensive processing-policy / transformation-rules reference artifact so consumers… Proposed 0 #103
7 Capture submission case data (aggregate + case-level) as EvidenceLines/EvidenceItems — Submissions sometimes include case data — aggregate case-related values, and sometimes grouped or individual case-level evidence records. These are currently n… Proposed 0 #104
8 Capture functional-data submissions (incl. MaveDB) from ClinVar — ClinVar has long allowed functional data submissions alongside regular classification SCVs, and has recently improved UI access and display of this data and ad… Proposed 0 #105
9 Backfill historical ClinVar releases (potentially to July 2019) — Currently, ClinVar-GKM releases start from July 2026. We could backfill historical releases as far back as July 2019 — when ClinVar's VCV XML files were first … Proposed 0 #107
10 Temporal representation of ClinVar data for analytics — Provide a temporal representation of ClinVar data to support analytics on how the data changes over time — e.g. the changing nature of classifications, the rat… Proposed 0 #109
11 Submitter tool: compare, manage, and simulate ClinVar submissions (with API) — Provide a submitter tool that works with these data releases to help submitters: Proposed 0 #110

How it works

  • Upvote: open an idea's discussion and click the Upvote button (the ▲ in the bottom-left corner of the discussion's opening post). The upvote tally drives the ordering above. GitHub Discussions support upvotes only — there is no downvote. Add a comment with your use case for extra signal.
  • Propose a new idea: open a discussion in the Ideas category. If the maintainers authorize it (by adding the roadmap authorized label), it appears in the table automatically.
  • Follow progress: an idea's status is updated as it moves toward implementation; the linked discussion carries the detailed conversation.

Feedback and feature requests are always welcome via the GitHub issue tracker as well.