ATLAS OF GENETIC RELATIONSHIPS
Every card answers five questions: what is linked, how reliable it is, whether it can be measured, whether it changes a decision, and where knowledge ends.
Sources and links are editorially checked. Every relationship reports evidence, actionability, testing method and the boundary of interpretation separately.
LDLR
Pathogenic variants can impair clearance of LDL cholesterol from blood.
Limit: Not every high LDL level is caused by LDLR; a VUS does not confirm a diagnosis.
Open relationCFTR
Two disease-causing CFTR variants, usually on opposite gene copies (in trans), can cause cystic fibrosis; manifestations depend on the exact variants.
Limit: One variant often indicates carrier status, not disease; phenotype and variant phase matter.
Open relationHBB
Clinically significant HBB variants alter the structure or amount of beta-globin.
Limit: A raw SNP list without haemoglobin and clinical context is insufficient.
Open relationHTT
CAG repeat expansion above clinical thresholds is associated with disease.
Limit: Testing may predict high disease risk, but not exact onset age or individual course.
Open relationBRCA1
Pathogenic variants increase risks of several cancers, especially breast and ovarian cancer.
Limit: A positive result does not make cancer inevitable; a negative result does not erase population or family risk.
Open relationBRCA2
Pathogenic variants raise risks of several cancers in people of different sexes.
Limit: Risks depend on sex, age, family history and the exact variant.
Open relationPALB2
Pathogenic germline variants raise breast-cancer risk and can matter in other cancers, including pancreatic cancer.
Limit: A tumour variant does not prove germline origin, and a VUS is not a basis for preventive surgery or targeted treatment.
Open relationMLH1
Pathogenic germline variants impair mismatch repair and raise risks of colorectal, endometrial and several other cancers.
Limit: Tumour loss of MLH1/PMS2 is not synonymous with Lynch syndrome; sporadic MLH1 promoter hypermethylation is possible.
Open relationMSH2
Pathogenic MSH2 variants, and some EPCAM deletions, can disable mismatch repair and raise risks of several cancers.
Limit: Tumour MSI-H/dMMR is a treatment biomarker but does not by itself prove an inherited syndrome.
Open relationMSH6
Pathogenic germline variants raise Lynch-associated cancer risks, with a spectrum and age profile distinct from MLH1/MSH2.
Limit: Normal IHC does not exclude every pathogenic variant; a VUS does not confirm the syndrome.
Open relationPMS2
Pathogenic germline variants raise cancer risk, generally with penetrance different from MLH1/MSH2.
Limit: A simplified panel or exome may cover some PMS2 exons unreliably.
Open relationTP53
Constitutional pathogenic TP53 variants are linked to a broad tumour spectrum and early onset.
Limit: A low TP53 variant fraction in blood does not always mean an inherited syndrome; variant origin must be resolved.
Open relationCDH1
Pathogenic germline variants are associated with diffuse gastric and lobular breast cancer risks.
Limit: Not every gastric cancer or CDH1 variant belongs to this syndrome.
Open relationPTEN
Pathogenic germline variants are associated with hamartomas and increased risks of several cancers.
Limit: Somatic PTEN loss in a tumour is not equivalent to a germline syndrome.
Open relationSTK11
Pathogenic germline variants are linked to characteristic polyps, pigmentation and increased tumour risks.
Limit: A somatic STK11 mutation in lung cancer has a different meaning and does not confirm Peutz–Jeghers syndrome.
Open relationRET
Activating germline RET variants are linked to medullary thyroid cancer and MEN2; risk depends on the exact codon.
Limit: A germline activating RET substitution and an acquired tumour RET fusion must not be conflated.
Open relationAPC
Pathogenic variants can cause familial adenomatous polyposis and related phenotypes.
Limit: A normal consumer SNP test does not rule out the syndrome.
Open relationPKD1
Pathogenic variants are a common cause of ADPKD; testing is technically challenging because of pseudogenes.
Limit: Not every panel or exome assay covers PKD1 reliably.
Open relationFBN1
Pathogenic FBN1 variants are associated with a spectrum of connective-tissue features.
Limit: Severity can vary substantially even within a family.
Open relationTTR
Pathogenic variants can cause amyloid polyneuropathy and/or cardiomyopathy.
Limit: Onset and organ involvement vary; carrier status is not the same as current disease.
Open relationCYP2C19
Some alleles reduce formation of clopidogrel’s active metabolite.
Limit: Do not change medicine or dose yourself; clinical context matters beyond genotype.
Open relationSLCO1B1
Transporter function affects exposure to some statins and risk of muscle symptoms.
Limit: It does not explain all muscle symptoms or replace assessment of drug interactions.
Open relationHLA-B
HLA-B*57:01 is associated with high risk of abacavir hypersensitivity.
Limit: A negative result does not rule out every possible drug reaction.
Open relationTPMT
Reduced TPMT activity increases risk of severe myelosuppression at standard thiopurine doses.
Limit: It does not replace NUDT15 assessment, interaction review or clinical monitoring.
Open relationNUDT15
Reduced NUDT15 function is associated with thiopurine intolerance and is particularly relevant in several populations.
Limit: A limited panel can miss rare functional variants.
Open relationDPYD
Reduced DPD activity can increase risk of severe fluoropyrimidine toxicity.
Limit: A common-variant panel does not detect every cause of DPD deficiency.
Open relationHERC2
Regulatory variants near OCA2 explain a substantial share of eye-colour variation in studied populations.
Limit: Accuracy depends on ancestry and model; one SNP does not describe a person’s appearance.
Open relationMCM6
Regulatory variants in MCM6 are associated with adult lactase persistence in some populations.
Limit: It does not diagnose every cause of dairy intolerance.
Open relationMC1R
Several functional variants substantially shift eumelanin–pheomelanin balance and the probability of red hair in several populations.
Limit: A variant alone does not determine hair shade, skin tone, freckle count or individual skin-cancer risk.
Open relationABCC11
Variant rs17822931 has an unusually strong effect on dry/wet earwax and production of axillary odour precursors.
Limit: Hygiene, skin microbiome and odour perception are not determined by one variant.
Open relationTAS2R38
PAV/AVI haplotypes substantially affect perception thresholds for PTC and PROP.
Limit: Coffee, vegetable or sweet preferences depend on culture, experience, age and many receptors.
Open relationALDH2
Reduced-function variant rs671 can cause acetaldehyde accumulation, flushing and unpleasant symptoms after alcohol.
Limit: Absence of rs671 does not make alcohol safe or exclude flushing from other causes.
Open relationACTN3
R577X has been studied for power and sprint traits, but effects are small, heterogeneous and do not determine talent.
Limit: A commercial ‘sprinter’ or ‘endurance’ label exceeds the evidence.
Open relationAPOE
APOE alleles modify relative risk and age at onset but do not determine a person’s fate.
Limit: Risk depends on age, ancestry, family and vascular context; testing minors without indication is unethical.
Open relationFOXO3
Some variants have repeatedly associated with longevity, but effects are small and population-dependent.
Limit: A small group-level association cannot be translated into a promise of extra years.
Open relationADDITIONAL MODULE
A separate map shows where genetics is clinically meaningful and where ability or talent tests exceed the evidence. It includes a dedicated children’s section.
Open module →