The Principle
Natural selection is the process by which individuals with certain heritable traits tend to survive and reproduce more successfully than others. At the genetic level, this translates to changes in allele frequencies over generations.
Fitness and Selection
Fitness (w) measures the relative reproductive success of a genotype. By convention, we set the highest fitness to 1 and express others relative to it.
Where:
- s = selection coefficient (reduction in fitness of aa)
- h = dominance coefficient (0 = recessive, 0.5 = codominant, 1 = dominant)
Change in Allele Frequency
The change in allele frequency per generation depends on current frequencies and fitness values:
Where w̄ = p²w_AA + 2pqw_Aa + q²w_aa is the mean population fitness.
Interactive Selection Dynamics
Adjust the fitness values below to see how selection changes allele frequencies. The curve shows Δp for each frequency p. Where the curve crosses zero, populations are at equilibrium.
Δp curve showing direction and magnitude of selection. Arrows indicate direction of change.
Selection Regimes
Directional Selection
When one allele has consistently higher fitness, selection drives it toward fixation. Try setting wAA = 1.0, wAa = 0.9, waa = 0.8 to see directional selection favoring A.
Balancing Selection
When heterozygotes have the highest fitness (overdominance), both alleles are maintained at a stable equilibrium. Try wAA = 0.9, wAa = 1.0, waa = 0.8.
Disruptive Selection
When heterozygotes have the lowest fitness (underdominance), the equilibrium is unstable — populations evolve away from it toward fixation or loss.
Rate of Selection
How fast does selection work? The time to change frequency depends on dominance:
- Dominant advantageous: Fast initial spread, slow to fix (rare recessives hard to remove)
- Recessive advantageous: Slow initial spread (hides in heterozygotes), fast final fixation
- Additive: Relatively constant rate throughout
Selection and Drift
In finite populations, selection competes with drift. The key parameter is Ns:
- Ns >> 1: Selection dominates; advantageous alleles likely fix
- Ns ≈ 1: Both forces matter; outcomes are probabilistic
- Ns << 1: Drift dominates; alleles behave as if neutral
The probability of fixation for a new mutation with selective advantage s in a population of size N:
Mutation-Selection Balance
Deleterious alleles persist in populations due to the balance between mutation (introducing them) and selection (removing them):
This explains why many genetic diseases persist at low frequencies — they're continuously replenished by mutation even as selection removes them.