Variant evidence review
Do not collapse ClinVar assertions into an automated classification
ClinVar aggregates submitted interpretations and their review status. An automated ACMG/AMP result is a different artifact: it applies a published evidence contract at a particular time. Both require provenance and qualified review.
Agent question
What does ClinVar report, what does the automated result report, and where do they agree or differ?
Compare provenance, not only labels
Preserve each ClinVar assertion, submitter context, review status and dates where returned. Then separately report the current automated classification, applied criteria and evidence versions.
Handle discordance explicitly
A disagreement is a review signal, not proof that either side is wrong. Differences can arise from evidence dates, access to case-level data, gene-specific specifications or criterion application.
- Variant identity and transcript
- ClinVar aggregate and individual assertions
- Review status and assertion dates
- Automated classification and applied criteria
- Evidence source versions
- Unresolved conflicts and missing data
Public references
Clinical and data boundary
Folklore Clinical Variant Interpretation MCP is published by Helena Bioinformatics. It accepts public variant-level queries only, not patient, phenotype, family, segregation or private case data. Its results are automated decision support for qualified professional review, not a diagnosis or treatment recommendation.