Specialized findings

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.

Eight of these condition-specific modules run automatically but don't have their own dashboard page — their findings appear in the Findings Explorer. Each looks at a small, well-defined set of variants. G6PD and BChE are read-only, on-demand API context modules. Neither is part of the standard analysis or stores a Findings Explorer entry; their sections below explain how to request the API summaries directly.

Hereditary haemochromatosis

Genes/variants: HFE C282Y (rs1800562), H63D (rs1799945). Classifies your HFE genotype (e.g. C282Y homozygous, compound heterozygous) for hereditary iron overload, with sex-stratified penetrance text. Most people with these genotypes are never diagnosed (≈88% of men, ≈97% of women with C282Y/C282Y in one large cohort), and compound-heterozygous calls are phase-inferred from unphased array data.

Inherited thrombophilia

Genes/variants: Factor V Leiden (F5 rs6025), Prothrombin G20210A (F2 rs1799963). Reports your combined genotype with odds ratios and absolute-risk context (risk rises mainly around triggers such as hormones, pregnancy, surgery, or immobility). This is a relative-risk module — absolute lifetime risk for most carriers stays low, and asymptomatic carriers are not routinely anticoagulated.

Alpha-1 antitrypsin deficiency

Gene/variants: SERPINA1 Pi*Z (rs28929474), Pi*S (rs17580). Reports PiZZ / PiSZ / PiSS / PiMZ / PiMS combinations with smoking and clinical-context notes. The array types only Pi*Z and Pi*S, so rarer deficiency alleles aren't detected, and PiSZ is phase-inferred.

Age-related macular degeneration (AMD)

Genes/variants: CFH Y402H (rs1061170), ARMS2/HTRA1 (rs10490924). Reports common risk-allele combinations with odds ratios and absolute-risk caveats. These are common GWAS variants (not pathogenic mutations), so evidence is capped; actual AMD risk also depends on age, smoking, and ~50 other variants.

APOL1 kidney risk

Gene/variants: APOL1 G1 (rs73885319, rs60910145), G2 (rs71785313), with an N264K modifier. A recessive-style risk model (two risk alleles = high risk) validated primarily in African-ancestry populations; results are ancestry-gated and partial genotypes are never reported as falsely low-risk.

Gout & serum urate

Genes/variants: ABCG2 Q141K (rs2231142), SLC2A9 (rs13129697). Reports urate-transporter risk genotypes with ancestry-stratified odds ratios (the ABCG2 effect is larger in East-Asian ancestry). Gout is multifactorial and most carriers never develop it; the module gives no dietary or treatment advice.

Leber hereditary optic neuropathy (LHON)

Variants: mtDNA m.11778G>A, m.3460G>A, m.14484T>C. Reports each of the three primary LHON mutations if detected. Penetrance is incomplete and sex-biased (≈50% of male vs ≈10% of female carriers ever lose vision), inheritance is maternal only, and the array cannot measure heteroplasmy (the fraction of affected mtDNA).

MT-RNR1 (aminoglycoside ototoxicity)

Variants: mtDNA m.1555A>G, m.1494C>T, m.1095T>C. Following CPIC's aminoglycoside-avoidance guideline, reports carriers of these 12S-rRNA variants, which raise the risk of hearing loss from aminoglycoside antibiotics. Maternally inherited; heteroplasmy isn't measured; often off-chip, so a negative doesn't rule it out.

G6PD deficiency

Gene: G6PD — 13 CPIC-defined deficiency variants (A−, Mediterranean, Canton, and others). An X-linked, sex-aware pharmacogenomic context module: it assigns a phenotype (normal / variable / deficient / indeterminate) and lists high-risk oxidative drugs (e.g. rasburicase, primaquine, dapsone). Array coverage of these variants varies.

G6PD is not part of the standard analysis and does not store a Findings Explorer entry. Request its summary directly with GET /api/analysis/g6pd?sample_id=<id>; the app has no dedicated G6PD page.

A negative result depends on how much of the panel was actually read. If any of the curated deficiency variants could not be called, the result is reported as indeterminate with medication risk undetermined, rather than as a normal, risk-clearing result — a partial panel is not a negative panel, however many of its loci were read. Even a fully covered negative reads as "no tested deficiency allele detected" rather than "G6PD normal": over 200 deficiency variants are known, limited-variant genotyping misses carriers,² and only an enzyme-activity assay establishes G6PD status.¹ A detected deficiency allele stays actionable at any coverage — absence of evidence is not evidence of absence, but presence is presence.

BChE (butyrylcholinesterase)

Gene/variants: BCHE rs1799807 (atypical), rs1803274 (K-variant). Context-only background on sensitivity to the anaesthetic muscle-relaxants succinylcholine and mivacurium. BChE is not part of the standard analysis and does not store a Findings Explorer entry. Request its summary directly with GET /api/analysis/bche?sample_id=<id>; the app has no dedicated BChE page. It types only two of many BCHE variants and is purely interpretive background; true BChE deficiency is confirmed by an enzyme-activity assay with your anaesthesia team.

---

1. Expanded Clinical Pharmacogenetics Implementation Consortium Guideline for Medication Use in the Context of G6PD Genotype (Gammal et al., 2023, Clinical Pharmacology & Therapeutics 113(5):973-985; PMID 36049896; PMCID PMC10281211; accessed 2026-07-30) provides G6PD genotype as part of the diagnosis of G6PD deficiency and directs that high-risk medications be avoided in deficient individuals. ↩

2. Functional analysis of G6PD variants associated with low G6PD activity in the All of Us Research Program (Powell et al., 2024, Genetics 228(4); PMID 39607789; PMCID PMC11631396; accessed 2026-07-30) reports that 13% of All of Us participants carrying deficiency-causing variants would be missed by genotyping only c.202G>A, and that over 800 G6PD variants remain of uncertain significance. ↩