ClinVar Benchmark Dataset for Genomic Variant Interpretation

A benchmark dataset derived from ClinVar, a public archive of human genetic variants and their clinical significance. The dataset is hosted on figshare and was authored by Nathan Fortier, last updated on May 13, 2026. It is available as a 337.2 KB ZIP file under a CC-BY-4.0 license.

Use Cases

• Benchmarking variant pathogenicity prediction models based on ClinVar annotations.
• Training machine learning classifiers for clinical significance based on genomic features.
• Evaluating the consistency of variant interpretations across different submitters to ClinVar.

Strengths

• Dataset is small (337.2 KB), facilitating quick download and local processing.
• Clear provenance linking to the authoritative ClinVar database and a bioRxiv preprint.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for large-scale training.

Provenance

Source

ClinVar public database, supplementary material for a bioRxiv preprint.

Freshness

Last updated 2026-05-13 17:42:54.