Interpreting genetic variants: classes, evidence and databases

Finding a DNA change is the easy part; deciding whether it causes disease uses population data, prediction, family studies and shared databases.

Updated Nov 13, 2026

Key facts

Variants are classed in five tiers: pathogenic, likely pathogenic, uncertain significance (VUS), likely benign and benign (ACMG/AMP standard).
Evidence comes from population frequency, the predicted effect on the protein, functional studies, segregation with disease in the family, de novo occurrence and previous reports.
A variant that is more common in healthy people than the disease allows is very unlikely to cause it; population databases such as gnomAD give these frequencies.
A VUS must not guide medical decisions; it is followed up and may be reclassified, most often as benign.
Shared databases: ClinVar (interpretations submitted by laboratories), gnomAD (frequencies), OMIM (genes and disorders), DECIPHER (copy-number changes).

Exome or genome analysis filters thousands of variants by quality, rarity, predicted effect, the inheritance pattern that fits the family (de novo, biallelic, X-linked) and how well the gene matches the patient's features.
Most reference data come from people of European ancestry, so people from under-represented populations, including Latin Americans, receive more uncertain results and sometimes wrong classifications.
Interpretation changes as knowledge grows; laboratories reanalyse old results, and patients should know that a report can be updated.
Pathogenic means the variant can cause the disease; whether a given person develops it still depends on penetrance.

Also called: Variant of uncertain significance (VUS), ACMG classification, ClinVar and gnomAD

A BRCA2 VUS does not justify risk-reducing surgery; management follows the personal and family history.
Testing relatives to see whether a variant tracks with the disease can help reclassify it.

Think of the answer, then tap the question to see it.

What are the five classes of a genetic variant?

Pathogenic, likely pathogenic, uncertain significance, likely benign and benign.

How do population databases help interpret a variant?

A variant too frequent in healthy people for the prevalence of the disease is unlikely to be its cause.

Why do people of under-represented ancestry get more uncertain results?

Reference databases contain fewer people like them, so rare-but-harmless variants in their populations look unusual and uncertain.

Educational information, not medical advice. For any symptom, see a health professional.

Spotted a mistake?We are always improving for the community. Tell us what is wrong and our team will review it.Report a mistake

Representative model, for educational purposes only.