Chance and populations · quick reference
Selection meets chance
Finite populations are samples, not averages. A useful variant can vanish by luck; a mildly harmful one can spread. Genomics made this population-genetic caveat impossible to ignore.
Short version: inheritance is a sample. In a small population, luck can make a gene version common or erase it—even when it does not help or harm.
Four forces leave different fingerprints
- Mutation creates sequence variants.
- Gene flow moves variants between populations.
- Selection biases reproduction because variants differ in effects.
- Drift changes frequencies through chance sampling, strongest in small populations.
Allele frequency is the share of sampled gene copies carrying a variant—for example, 20 A copies among 40 total gives 0.5. See the field guide whenever a term is unfamiliar.
Effective population size (Nₑ) is not headcount. Bottlenecks, unequal family sizes, sex ratios, and population structure can make it much smaller. A rough intuition: when selection is weak relative to about 1/Nₑ, drift can dominate its fate.
Simulation: same rules, twelve histories
Every line starts with the same 50/50 mix. Run the model and watch identical starting populations diverge.
Optional: model assumptions
This deliberately simple haploid Wright–Fisher model samples N gene copies each generation. The selection coefficient s changes A’s relative reproduction (positive favors A); migration m is the fraction replaced by migrants from a population where A has frequency 0.8. Dominance, mutation, linkage, and population structure are omitted.
| Replicate | Gen 0 | Gen 10 | Gen 20 | Gen 30 | Gen 40 |
|---|
Try: N=20 vs 500 with selection 0; then set selection to 0.05. Selection changes the tendency, but chance still changes each history.
The neutral and nearly neutral update: Kimura and Ohta’s theories predate 1995, but genome-scale data made neutrality a standard null model. Many molecular changes are neutral or weakly selected; that does not mean organisms are random or adaptation is unreal. Neutrality is a claim about a particular variant in a particular population context.
Try the case
An island is founded by five randomly carried lizards. A rare color becomes common even though colors survive equally well. Best explanation?
Go deeper
- Charlesworth (2009), “Effective population size and patterns of molecular evolution”Nature Reviews Genetics · expert review.
- Kern & Hahn (2018), “The Neutral Theory in Light of Natural Selection”A modern review and argument; useful precisely because proportions remain debated.
- Eyre-Walker & Keightley (2007), distribution of fitness effectsHow real mutations span strongly harmful through effectively neutral.