ClinVar

Key facts

Maintained by

NCBI, National Library of Medicine, NIH

Record types

SCV (submission), RCV (variant-condition), VCV (variant)

Review status scale

0 to 4 stars, based on submitter criteria and review level

Classification types

Germline, somatic clinical impact, oncogenicity (since 2024)

What is ClinVar?

ClinVar is where clinical laboratories, research groups and expert panels deposit their assessments of whether a genetic variant is pathogenic, benign or of uncertain significance for a condition. Each submission is kept as its own record with the submitter name, date, condition and the criteria used.

ClinVar is an archive, not a judge. NCBI validates and organizes submissions, but it does not evaluate the evidence behind most of them. When submitters disagree, ClinVar reports the conflict rather than choosing a side.

How ClinVar records and review stars work

Each submitted record (SCV) is aggregated into a variant-condition record (RCV) and a variant record (VCV). ClinVar computes an aggregate classification and a review status for these records from the contributing submissions. Review status is shown as stars:

• Four stars: practice guideline.
• Three stars: reviewed by expert panel.
• Two stars: criteria provided, multiple submitters, no conflicts.
• One star: criteria provided, single submitter; or criteria provided, conflicting classifications.
• Zero stars: no assertion criteria provided, or no classification provided.

Why ClinVar matters

ClinVar is the most widely used public reference for the clinical meaning of specific variants. Diagnostic developers use it when building evidence for tests, and researchers use it to estimate how many known pathogenic variants exist in a gene of interest.

Its variation IDs and HGVS descriptions make it a practical join key between genetic testing results, published literature and gene-level datasets such as MeSH-indexed research.

Limitations of ClinVar data

A classification in ClinVar reflects what submitters reported, when they reported it. Records can be outdated, zero-star records carry little documented evidence, and conflicting classifications are common for some genes. Filtering by review status and by classification type before counting variants avoids overstating certainty.