Glossary

Plain-language definitions for the terms used across your report. These describe concepts; they are not medical advice.

Variant: A place where your DNA sequence differs from the reference genome. Most variants are harmless; a few are studied in the clinical literature.
Reference genome (GRCh38): The standard human DNA sequence that everyone's data is compared against, so a position means the same thing for everyone. GRCh38 is the current build.
Zygosity: Whether a variant is present on one copy of a gene (heterozygous) or both copies (homozygous). For recessive conditions this is the difference between being a carrier and being affected.
Carrier: Someone with one copy of a recessive variant. Carriers are typically unaffected themselves, but the variant can matter for family planning.
Pathogenic / Likely pathogenic (P/LP): ClinVar's classification for variants the field considers disease-causing or probably disease-causing. We restate ClinVar's label; we do not assign our own.
Review status (gold stars): ClinVar's 0–4 star measure of how well-reviewed a classification is: more stars means more agreement among submitters. It answers "how sure is the field?"
Penetrance: How often people with a given variant actually develop the associated condition. High penetrance does not mean certainty.
Allele frequency: How common a variant is in the population (from gnomAD). Very rare variants are handled differently from common ones.
Genome-wide significance (p ≤ 5×10⁻⁸): The strict statistical bar a genome-wide association study (GWAS) has to clear before it will call a variant-and-trait link real. A GWAS tests roughly a million common variants at once, so the ordinary p < 0.05 cutoff would throw up tens of thousands of false hits by chance alone. Spreading that 0.05 across all those tests lands at about 5×10⁻⁸. Clear it, and the association is very unlikely to be a statistical fluke.
Below genome-wide significance: A result that did not clear the 5×10⁻⁸ bar. This is where readers are most often misled, so it is worth stating plainly: below the threshold means "not proven by this study on its own", not "false" and not "no effect". The bar is deliberately strict to keep false discoveries out of the headline list, and plenty of genuine associations sit just underneath it, often because a single study was simply too small to reach it even when later work confirms the effect.

We include some below-threshold results on purpose, and we always label them as such. There are two real reasons. First, a polygenic score (PRS) is built by adding up many small-effect variants, and a lot of them fall below 5×10⁻⁸ one at a time; dropping them would throw away real predictive signal, because the threshold is a discovery filter, not a verdict on truth. Second, the Frontier and Fringe sections exist precisely to surface emerging, contested, or single-study findings with their evidence strength stated up front, so you can see the lively edge of the science without us dressing a lead up as a settled fact.
Pharmacogenomics (PGx): How your genetics may affect your response to certain medications, based on CPIC guidance. Always discuss medication decisions with a clinician.
Haplogroup: A branch of the human family tree defined by shared inherited markers. The maternal line is traced through mitochondrial DNA.
Literature match: A report finding: your data matches a variant described in a public database. It is information to confirm with a clinician, not a diagnosis.

See how the analysis works for how these fit together.