2.11 Human Genetic Variation & Disease Susceptibility
Key Takeaways
- Human genomic variation includes SNVs, indels, CNVs, and larger structural variants; clinical impact depends on size, gene content, and frequency context.
- Common variants (often MAF ≥1% or ≥5% by study definition) and rare variants play different roles in disease architecture and in how counselors interpret reports.
- GWAS associate common variants with traits/diseases, usually tagging loci via linkage disequilibrium rather than proving a single causal mutation.
- Polygenic risk scores aggregate many small-effect variants into an individualized relative-risk estimate but are ancestry-sensitive and usually not deterministic diagnoses.
- CGC counseling must separate diagnostic Mendelian findings from susceptibility/GWAS/PRS results and communicate absolute risk, uncertainty, and limits of clinical utility.
Why variation and susceptibility show up on Domain 2B
After you can calculate HWE carrier risks, the outline still expects you to explain what kinds of DNA differences exist, how common versus rare variation relates to disease, and how genome-wide association studies (GWAS) and polygenic risk scores (PRS) inform—but do not replace—Mendelian counseling. Clients increasingly arrive with consumer genomics printouts, PRS from research or specialty clinics, or reports that mix diagnostic panel results with “risk alleles.” Your job is to classify the finding and counsel accordingly.
Major types of human genetic variation
| Variant class | Typical description | Board-relevant notes |
|---|---|---|
| SNV (single-nucleotide variant) | One base substituted for another | Includes common SNPs and rare pathogenic SNVs; most GWAS hits are SNVs |
| Indel | Small insertion or deletion (often <50 bp in informal clinical speech) | Frameshifting indels in coding sequence can be highly disruptive |
| CNV (copy-number variant) | Gain/loss of a stretch of DNA (kb to Mb scale) | Dosage-sensitive genes; microarray/exome CNV calling; can be benign polymorphisms or pathogenic |
| Structural variant (SV) | Inversions, translocations, large deletions/duplications, complex rearrangements | May be cytogenetically visible or cryptic; gene disruption or fusion |
| Repeat expansion | Pathologic expansion of short tandem repeats | Special class (for example, Fragile X); not a standard GWAS feature |
SNP vs SNV wording: SNP historically implied a common polymorphism; SNV is the broader term for any single-base change regardless of frequency. On the exam, focus on frequency and clinical assertion, not pedantic label wars—but know that a “SNP chip” primarily assays common SNVs.
Common versus rare variation
| Category | Typical frequency language | Role in disease architecture |
|---|---|---|
| Common | Minor allele frequency (MAF) often ≥1% or ≥5% (threshold varies by study) | Small effect sizes individually; collectively important for complex traits; backbone of GWAS/PRS |
| Low-frequency / rare | MAF <1% down to private/family-specific | More likely to include high-impact Mendelian alleles; burden tests and diagnostic sequencing focus here |
| Private | Seen in one individual/family | May be pathogenic, VUS, or benign—needs ACMG-style interpretation (Domain 3), not GWAS logic |
Counseling pearl: A common GWAS risk allele with odds ratio 1.1 is not analogous to a pathogenic BRCA1 truncation. Frequency alone does not equal benignity for rare variants, and rarity alone does not equal pathogenicity—but the tools and language differ.
From variation to susceptibility: GWAS concepts
A genome-wide association study genotypes (or imputes) hundreds of thousands to millions of common variants across many people and tests which alleles associate with a trait or disease more often than expected by chance.
Core GWAS vocabulary for counselors
| Term | Meaning in practice |
|---|---|
| Case–control or cohort design | Compare allele frequencies in affected vs unaffected (or correlate with quantitative traits) |
| Genome-wide significance | Very stringent p-value threshold (commonly ~5 × 10⁻⁸) to account for multiple testing |
| Linkage disequilibrium (LD) | Non-random correlation of nearby alleles; the associated SNP may be a tag rather than the causal variant |
| Manhattan plot | Visual of −log₁₀(p) by genomic position; “skyscrapers” mark associated loci |
| Odds ratio / effect size | Strength of association; complex-trait ORs are often modest (near 1.1–1.5 per locus) |
| Ancestry matching / stratification control | Prevents false associations driven by ancestry differences rather than disease biology |
What GWAS does not automatically give you
- A complete genetic diagnosis for an individual
- Proof that the lead SNP is the functional mutation
- Effect sizes that transfer perfectly across ancestries
- Replacement for family history, empiric risks, or Mendelian testing when indicated
Exam scenario: A client says, “My 23andMe report says I have a SNP linked to type 2 diabetes.” Best counseling frame: that finding is a common-variant association contributing a small relative risk, interpreted with ancestry, lifestyle, and family history—not a deterministic disease gene result.
Polygenic risk scores (PRS): mechanics and counseling caveats
A polygenic risk score (also called polygenic score) combines information from many variants—often thousands to millions—into one number estimating relative genetic predisposition for a trait:
Conceptual formula: PRS ≈ Σ (effect weightᵢ × allele countᵢ)
over selected variants, using weights from GWAS summary statistics (sometimes with Bayesian or LD-adjusted methods).
What PRS can do (appropriately framed)
- Stratify a population into higher vs lower relative risk tails for a complex disease
- In some clinical research or specialty pathways, refine screening age/intensity discussions when guidelines support it
- Complement—not erase—family history and clinical risk models
Hard caveats genetic counselors must voice
| Caveat | Why it matters in session |
|---|---|
| Ancestry / portability | Most GWAS discovery cohorts are European-ancestry biased; PRS performance often degrades in other ancestries |
| Not diagnostic | High PRS ≠ disease; low PRS ≠ immunity |
| Absolute vs relative risk | Clients hear “high risk” and imagine Mendelian certainty; convert to absolute risk ranges when data allow |
| Environmental & clinical context | BMI, smoking, age, screening history may dominate or modify genetic risk |
| Laboratory & method heterogeneity | Different variant sets/weights → different scores; scores are not interchangeable across vendors |
| Psychosocial impact | Anxiety, false reassurance, cascade confusion with relatives who share only part of the polygenic background |
| Equity | Offering PRS without explaining ancestry limits can worsen disparities in perceived benefit |
Worked counseling contrast (numbers for intuition)
Suppose population lifetime risk for a disease is 10%. A PRS places someone at 1.5× average genetic risk versus the middle of the distribution.
- Naive relative framing: “50% higher risk than average.”
- Absolute framing: roughly on the order of ~15% lifetime risk if the multiplier applies cleanly to lifetime risk (real models are disease-specific and age-dependent—do not invent precision beyond the stem).
Compare that with a high-penetrance Mendelian pathogenic variant that might confer lifetime risks of 40–80%+ for certain cancers: the communication register must differ.
Integrating variation classes into susceptibility counseling
Use this decision scaffold on stems and in clinic:
- What variant class was reported? (pathogenic SNV on a diagnostic panel vs common GWAS SNP vs CNV of uncertain significance)
- What is the intended use? (diagnose a Mendelian condition, refine complex-disease risk, research only)
- What is the frequency context? (rare family variant vs common polymorphism)
- What is the ancestry context? (does the evidence base match the client?)
- What decisions change? (surveillance, testing relatives, reproductive options, lifestyle)—if none, say so clearly
Table — matching result type to counseling stance
| Result type | Typical stance |
|---|---|
| Pathogenic/likely pathogenic Mendelian variant | Diagnostic/ predictive counseling; cascade testing; guideline-driven management |
| VUS on diagnostic test | Uncertainty counseling; avoid predictive use; may reclassify later |
| Common GWAS risk alleles / consumer SNP reports | Susceptibility education; modest effects; lifestyle + family history |
| Clinical PRS (when offered) | Relative/absolute risk education; ancestry limits; not a substitute for indicated Mendelian testing |
| Pathogenic CNV at dosage-sensitive locus | Often Mendelian/ contiguous-gene framing; recurrence depends on mechanism (de novo vs inherited) |
Exam scenarios and traps
- Trap — equating GWAS SNP with disease-causing mutation: Association ≠ causation at the tagged site; effect sizes are usually small.
- Trap — offering the same PRS interpretation across ancestries: Portability limits are a standard correct answer theme.
- Trap — telling a low-PRS client they need no screening: PRS does not erase population screening guidelines or strong family history.
- Trap — calling every indel “pathogenic”: Impact depends on reading-frame, gene, and classification evidence.
- Scenario: A prenatal couple brings a microarray showing a common polymorphic CNV listed as benign in population databases—counsel as likely benign population variation, not as unexplained disease etiology, unless new evidence suggests otherwise.
- Scenario: An oncology patient with a strong hereditary cancer pedigree receives a negative multigene panel plus a “elevated breast cancer PRS.” Next best emphasis: PRS does not explain the Mendelian-looking family history; consider phenotype-driven differentials, undetected variants, phenocopies, or other genes—PRS is adjunct susceptibility information only.
Synthesis with population genetics (Domain 2B bridge)
HWE carrier math primarily serves rare Mendelian recessive alleles in populations. GWAS/PRS primarily serve common variant contributions to complex traits. Founder populations can enrich rare pathogenic alleles and shift common-allele frequencies—so ancestry specificity matters in both frameworks. Strong CGC answers name which framework the stem is using before calculating or reassuring.
In a GWAS for a complex disease, a lead SNP reaches genome-wide significance with an odds ratio of 1.15. What is the most accurate counseling interpretation?
Which statement best describes a major limitation of many current polygenic risk scores in clinical counseling?
A copy-number gain of several hundred kilobases that changes gene dosage at a known dosage-sensitive locus is best classified as which variation type?
A client with a compelling autosomal dominant cancer pedigree receives a negative guideline-based multigene panel and a consumer report showing several common GWAS risk SNPs. What is the most appropriate counseling emphasis?