Intermediatepopulation
Genetic Drift
Overview
Genetic drift is the random fluctuation of allele frequencies due to sampling error in finite populations: by chance, some individuals leave more offspring, shifting frequencies unpredictably each generation. Drift is strongest in small populations, inevitably fixes or loses alleles over time, and reduces heterozygosity at a rate of 1/(2N) per generation. It is evolution without adaptation.
Learning objectives
- Define genetic drift and its cause (sampling error)
- Explain why drift is stronger in small populations
- Describe fixation, loss, and heterozygosity decay
- Distinguish drift from selection
Sampling error, writ large
Each generation samples 2N gametes from the parental gene pool. In a population of 10 with p = 0.5, the next generation's p might be 0.4 or 0.6 purely by chance; in a population of 10,000, it stays near 0.5. This binomial sampling variance, p(1-p)/2N, IS drift: bigger N, smaller wobble.
Variance of next generation's p is p(1-p)/2N: the formula shows exactly why small N means big drift.
Fixation and loss
Without other forces, drift eventually drives every allele to fixation (p=1) or loss (p=0); heterozygosity decays as H_t = H_0(1 - 1/2N)^t. Neutral alleles fix with probability equal to their current frequency. Most new mutations (starting at 1/2N) are lost by drift within a few generations.
It does not 'prefer' any allele, yet given time it eliminates all variation not maintained by other forces.
Founder effects and bottlenecks
Drift's dramatic cases: a few colonists carry unrepresentative allele frequencies (founder effect: Amish Ellis-van Creveld), or a catastrophe slashes N (bottleneck: cheetahs' near-zero variation, northern elephant seals). Both leave lasting signatures: reduced heterozygosity and shifted frequencies.
Two ancient bottlenecks left cheetahs so uniform that skin grafts between unrelated individuals are accepted.
Drift vs selection
Selection dominates when s >> 1/2N; drift dominates when s << 1/2N. In small populations, even beneficial alleles are often lost and deleterious ones fixed by chance. Conservation genetics fights drift with managed breeding maximizing effective population size.
Compare the selection coefficient to the drift threshold: the bigger force wins.
Key points
- Drift = random allele-frequency change from finite sampling
- Strength scales as 1/2N: small populations drift hard
- Drift fixes or loses alleles; heterozygosity decays (1-1/2N)^t
- Founder effects and bottlenecks are drift's dramatic forms
Common misconceptions
✓ Truth: It acts in all finite populations; it is just overshadowed by selection in large ones. Even N=10,000 drifts measurably over millennia.
✓ Truth: Selection is deterministic fitness differences; drift is random sampling. They are distinct, sometimes opposing, forces.
Practice questions
Q1. Genetic drift is strongest when population size is
- Large
- Small
- Infinite
- Growing
Sampling variance p(1-p)/2N grows as N shrinks.
Q2. Over long time spans, drift alone will
- Maintain all alleles
- Fix or lose every allele
- Increase heterozygosity
- Create new alleles
Random walks eventually hit the boundaries p=0 or p=1.
Q3. Cheetahs accept skin grafts from unrelated individuals because
- They are clones
- Ancient bottlenecks erased genetic variation
- They lack immune systems
- Grafts always work
Extreme drift through bottlenecks made cheetahs nearly genetically identical.
Click a question to reveal the answer and explanation.
Frequently asked questions
What is effective population size (Ne)?
The size of an idealized population drifting at the observed rate; skewed sex ratios, bottlenecks, and variance in offspring lower Ne below census N.
Can drift cause adaptation?
No: drift is random with respect to fitness. It can fix deleterious alleles, the opposite of adaptation.
How fast are new mutations lost?
About 1 - 1/2N per generation are lost almost immediately; only ~1/2N of new neutral mutations eventually fix.