Skip to content

Familial hypercholesterolemia

Not a medical test — for research and educational use only

Yeliztli analyses consumer genotyping-array data (23andMe / AncestryDNA), which is not a clinical-grade test. Results are not diagnostic, are not clinically validated, and must not be used to make medical decisions. Array data is especially unreliable for rare, disease-causing variants. Always confirm any finding with an accredited clinical laboratory and discuss it with a qualified clinician or genetic counsellor before acting on it. See the Intended use & disclaimers page for the details and the evidence behind this warning.

This view focuses on familial hypercholesterolemia (FH) — an inherited tendency toward very high LDL cholesterol — by combining three lines of evidence. It is framed against the clinical FH criteria (Dutch Lipid Clinic Network, Simon Broome) for educational reference only, and is not a clinical FH diagnosis.

What it looks at

  1. Monogenic FH variants in LDLR, APOB, and PCSK9 (read from the cardiovascular module's findings). APOB findings follow that module's existing FH criteria, including its ClinVar condition-label and lower-penetrance/risk-allele filters.
  2. APOB R3527Q (rs5742904), the familial-defective-apoB variant, if it's typed and you carry it.
  3. An LDL-C polygenic score (PGS000688) summarising many small-effect variants.

What you'll see

A three-part assessment: any monogenic FH variants you carry, your APOB R3527Q status, and your LDL-C polygenic burden. The polygenic part shows a raw score and how much of it your array actually covers. It also shows where the score comes from: the population it was developed in, the study's sample size, and how many variants it uses. The card's ancestry caveat refers to that development population.

It does not measure your cholesterol

This module reads genetics only — it does not measure your LDL-C or check for clinical signs of FH. The polygenic percentile is withheld when it can't be calibrated on un-imputed array data.

Good to know

  • It doesn't detect FH variants on its own — it summarises the cardiovascular module's findings plus the single APOB variant and the polygenic score.
  • APOB is bidirectional. APOB findings with low-LDL-only ClinVar labels—and protein-truncating APOB variants linked to familial hypobetalipoproteinemia even when ClinVar labels aggregate both directions—are reported under Other lipid metabolism and excluded from FH status. Other APOB findings follow the cardiovascular module's existing FH criteria, including its ClinVar condition-label and lower-penetrance/risk-allele filters. Human genetic studies associate protein-truncating variants with low LDL cholesterol (PMID:30939045; DOI:10.1161/CIRCGEN.118.002376 (accessed 2026-07-31); PMID:36723951; DOI:10.1001/jamacardio.2022.5271 (accessed 2026-07-31)).
  • PCSK9 is bidirectional too. FH-causing PCSK9 variants act by gain of function, so PCSK9 nonsense and frameshift variants are reported under Other lipid metabolism and excluded from FH status, even when ClinVar labels them as FH. Other PCSK9 findings follow the cardiovascular module's existing FH criteria. Human genetic studies associate nonsense and frameshift PCSK9 variants with low LDL cholesterol: nonsense variants (PMID:15654334; DOI:10.1038/ng1509 (accessed 2026-10-11); PMID:16989838; DOI:10.1016/j.atherosclerosis.2006.08.039 (accessed 2026-10-11)), a frameshift variant (PMID:17170371; DOI:10.1161/01.ATV.0000255311.26383.2f (accessed 2026-10-11)), and protein-truncating variants as a class (PMID:36723951; DOI:10.1001/jamacardio.2022.5271 (accessed 2026-10-11)). PCSK9 splice-site variants keep the existing FH criteria: studies count only splice changes that a stricter loss-of-function filter accepts, and a splice annotation alone does not show that.
  • High LDL cholesterol has many causes; genetics is only one. Discuss any concern, and your actual lipid panel, with a clinician.
  • The score is computed the first time you open this page for a sample and reused after that. It is computed again automatically when something it depends on changes (re-annotation or imputation of the sample, its inferred ancestry, the cardiovascular findings it summarises, the installed score database, reference data, or an app update), and Re-score computes it again on request. Computing it can take a few minutes.