2.10 Population Genetics, HWE & Quantitative Genetics
Key Takeaways
- Under Hardy–Weinberg equilibrium (HWE), allele frequencies satisfy p + q = 1 and genotype frequencies are p² (AA), 2pq (Aa), and q² (aa).
- For a rare autosomal recessive condition, disease incidence ≈ q², so the carrier frequency is approximately 2√(incidence) when q is small.
- HWE assumes random mating, no selection, no migration, negligible mutation, and a large population; founder effects, assortative mating, selection, and drift violate those assumptions.
- Quantitative (polygenic) traits reflect many loci of small effect plus environment; dichotomous multifactorial traits are often framed with a liability-threshold model.
- Board stems commonly ask you to estimate carrier risk from incidence, recognize when HWE does not apply to a subpopulation, or distinguish Mendelian vs polygenic risk framing.
Why population genetics matters on the CGC exam
Domain 2 asks you to move from a family pedigree to a population-informed risk. Carrier-screening panels, ethnicity-based residual risks, and pretest counseling for recessive conditions all rest on Hardy–Weinberg equilibrium (HWE) math and on knowing when that math fails. The same domain also expects board-level comfort with quantitative genetics—polygenic inheritance, multifactorial liability, and how those concepts differ from single-gene Mendelian calculations you used in earlier sections.
On exam day, expect stems that give an incidence (for example, 1 in 2,500 births) and ask for the approximate carrier frequency, or that describe an isolated community and ask why a published population carrier rate may not apply.
Hardy–Weinberg equilibrium: definitions
For a diallelic autosomal locus with alleles A (frequency p) and a (frequency q):
| Relationship | Formula | Meaning |
|---|---|---|
| Allele frequencies | p + q = 1 | All allele copies at the locus |
| Genotype frequencies at HWE | p² + 2pq + q² = 1 | AA, Aa, aa proportions in the population |
| Homozygous dominant | p² | Frequency of AA |
| Heterozygotes (carriers if recessive) | 2pq | Frequency of Aa |
| Homozygous recessive | q² | Frequency of aa (often ≈ disease incidence for AR conditions) |
Key counselor translation: when a rare autosomal recessive (AR) disease is fully penetrant and diagnosed at birth, observed incidence ≈ q². Solving for q = √(incidence) then yields the heterozygote (carrier) frequency 2pq. When the disease is rare, q is small, p ≈ 1, and 2pq ≈ 2q = 2√(incidence).
Worked example 1 — classic cystic fibrosis–style incidence
A Northern European population reports an AR disease incidence of 1/2,500 live births. Assume HWE, complete ascertainment, and that affected individuals are aa.
- q² = 1/2,500 → q = 1/50 = 0.02
- p = 1 − q = 0.98
- Exact carrier frequency: 2pq = 2(0.98)(0.02) = 0.0392 ≈ 3.92% ≈ 1/25.5
- Rare-disease approximation: 2q = 2/50 = 1/25 = 0.04 (4%)
Counseling use: “About 1 in 25 people in this reference population are carriers” is the approximation you will see in teaching materials; the exact 2pq is slightly lower because p is 0.98, not 1.
Worked example 2 — rarer AR condition
Incidence 1/40,000:
- q = √(1/40,000) = 1/200 = 0.005
- 2pq ≈ 2q = 1/100 = 1% (exact 2pq = 2 × 0.995 × 0.005 = 0.00995 ≈ 0.995%)
Worked example 3 — from carrier frequency back to allele frequency
A lab brochure states carrier frequency ≈ 1/50 for a rare AR condition under HWE.
- 2pq ≈ 1/50 = 0.02 → with p ≈ 1, q ≈ 0.01
- Expected affected incidence q² ≈ 0.0001 = 1/10,000
This reverse calculation helps you sanity-check whether a quoted carrier rate matches a quoted incidence.
HWE assumptions (memorize as a checklist)
HWE genotype frequencies hold when the population meets these ideal conditions:
| Assumption | Plain-language meaning | Clinical red flag if violated |
|---|---|---|
| Random mating | Mate choice independent of genotype at the locus | Consanguinity, assortative mating by phenotype/ethnicity |
| No selection | All genotypes contribute equally to the next generation | Lethal or fitness-reducing genotypes; heterozygote advantage |
| No migration | No gene flow into/out of the population | Admixture; recent immigrant subgroups |
| Negligible mutation | Allele frequencies not shifting from new mutations each generation | Dynamic mutation disorders (usually handled as special cases) |
| Large (effectively infinite) population | Sampling error does not dominate allele-frequency change | Small isolates → genetic drift and founder effects |
If assumptions fail, published pan-ethnic or large-population carrier rates may misstate risk for this consultand’s ancestry or community.
HWE violators the board loves
Founder effects
A founder effect occurs when a new population is established by a small number of individuals, so allele frequencies in descendants reflect the founders’ chance genotypes—not the source population. Classic teaching examples include elevated frequencies of certain AR alleles in Ashkenazi Jewish, French Canadian, Finnish, and other founder populations. Counseling implication: use population-specific carrier frequencies and panels when available, not generic U.S. averages.
Assortative mating
Assortative mating is non-random pairing by phenotype (for example, partners who both have hearing loss, short stature, or the same cultural/religious community). Positive assortative mating increases homozygosity relative to HWE expectations and can raise the chance both partners carry the same recessive alleles. Consanguinity is a related but distinct concept: mating between relatives increases the chance of sharing alleles identical by descent.
Selection
Natural selection changes allele frequencies across generations. Heterozygote advantage (for example, sickle cell trait and malaria resistance in endemic regions) maintains deleterious alleles at higher frequency than mutation–selection balance alone would predict. Negative selection against affected homozygotes reduces q over time unless balanced by mutation or heterozygote advantage. For counseling, selection explains why some AR alleles remain common; it also warns that incidence and carrier rates are population- and environment-specific.
Genetic drift
In small populations, allele frequencies fluctuate by chance (genetic drift). Drift can fix or lose alleles independent of fitness. Founder effects are a drift-related phenomenon at population founding. Drift is why tiny isolates can show unexpected disease frequencies.
Migration / population stratification
Gene flow and population stratification (systematic ancestry differences between groups being compared) matter for both HWE applications and association studies (next section). A “U.S. population” rate may blend groups with very different q values, producing a number that fits no single client’s ancestry well.
Quantitative genetics basics (board level)
Not every familial trait is single-gene Mendelian. Quantitative genetics addresses traits influenced by many loci (polygenic architecture) plus environmental factors.
| Concept | Board-level meaning | Counseling hook |
|---|---|---|
| Polygenic trait | Many variants of individually small effect contribute additively (or with interactions) | Height, many common complex diseases |
| Multifactorial | Genes + environment jointly influence risk | Neural tube defects, many congenital anomalies, common adult diseases |
| Heritability | Proportion of phenotypic variance in a population attributable to genetic variance | Does not mean “percent of this patient’s disease that is genetic” |
| Liability-threshold model | An unseen continuous “liability” distribution; disease appears when liability exceeds a threshold | Explains why recurrence risks rise with more affected relatives or more severe presentations |
| Polygenic risk (conceptual) | Aggregate burden of risk alleles | Bridges to polygenic risk scores (PRS) in the next section |
Liability-threshold model — how to explain it
Imagine a bell-shaped liability curve for a dichotomous multifactorial condition (affected vs unaffected). Everyone has a liability value; only those above a threshold are affected. Relatives of an affected person have a shifted liability distribution (shared genes ± shared environment), so more of them cross the threshold—hence empiric recurrence risks higher than population incidence. Recurrence risk typically increases when:
- More than one relative is affected
- The affected relative is more severely affected
- The affected relative is of the less frequently affected sex (for sex-biased thresholds)
- The consultand is a closer relative
These qualitative rules are high-yield for CGC stems even when no numeric table is provided.
Polygenic vs Mendelian framing — avoid the trap
| Feature | Mendelian (single-gene) risk | Quantitative / multifactorial risk |
|---|---|---|
| Core math | Segregation ratios, Bayes, HWE carrier math | Empiric risks, liability concepts, PRS (when used) |
| Typical counseling | Pathogenic variant → high relative risk | Modest risk shifts; population context essential |
| Family pattern | Vertical/horizontal patterns matching inheritance | Clustering without a clear single-gene pattern |
| Test implication | Diagnostic/ predictive single-gene or panel testing | Often lifestyle, screening, or research/PRS contexts |
Exam scenarios and traps
- Trap — using 2q when q is not small: For common “recessive” phenotypes with high q, use full 2pq, not 2q.
- Trap — applying a national carrier rate to a founder community: Prefer ancestry-specific data; founder effects violate the “same q as the large population” assumption.
- Trap — treating heritability as personal probability: Heritability is a population variance statistic, not an individual’s percent-genetic score.
- Scenario: Incidence 1/10,000 AR disease under HWE → q = 1/100; approximate carrier frequency 1/50. If the couple is from a known founder population with documented higher incidence, recalculate with the local incidence or use published local carrier rates.
- Scenario: Two partners with the same congenital anomaly seek recurrence counseling—consider assortative mating and shared multifactorial liability, not only a single Mendelian locus.
Practice checkpoints before you move on
- Convert incidence ↔ q ↔ 2pq fluently for rare AR disease.
- List the five HWE assumptions and map each major violator (founder effect, assortative mating, selection, drift, migration).
- Explain liability threshold in one sentence a client could follow.
- State why heritability ≠ “how genetic my child’s condition is.”
An autosomal recessive condition has an incidence of 1 in 10,000 in a large randomly mating population at HWE. What is the approximate carrier frequency using the rare-disease approximation?
Which situation most clearly violates the Hardy–Weinberg assumption of random mating?
Under the liability-threshold model for a multifactorial dichotomous trait, which observation typically raises recurrence risk for a consultand?
A counselor quotes a U.S. average carrier frequency for an AR disease to a client from a well-documented founder population where the disease is known to be more common. What is the best critique of that approach?