Genetics and evolution

What the exam asks

Expect a frequency to calculate, a described population change to attribute to a force, or a tree to read. The Hardy-Weinberg items are the ones candidates lose most reliably, and for two reasons. The first is reflex: the equation gets applied to any population the stem describes, including ones the stem has just told you are under selection or are tiny, where the answer is that the model does not hold and that is what is being tested. Check the five assumptions before you write p and q, and notice when the stem is handing you a violation on purpose. The second is 2pq. Asked for the frequency of carriers, candidates reach for q squared, which is the frequency of affected individuals. Carriers are heterozygotes, heterozygotes are 2pq, and the sanity check is that carriers of a rare recessive should come out far more common than sufferers. If your carrier figure is smaller than your affected figure, you have used the wrong term. On trees, the trap is proximity: two tips drawn side by side are not thereby relatives. Trace back to where the lines meet.

This topic is population thinking. An individual does not evolve; a population does, and what changes is the frequency of alleles in it. Every question here is ultimately about what pushes those frequencies around and what would happen if nothing did.

That last clause is what Hardy-Weinberg is for. It is a null model, not a law of nature: it says that with no mutation, no gene flow, no selection, random mating and a population large enough that chance does not matter, allele frequencies do not change and genotype frequencies settle at p squared, 2pq and q squared. No real population satisfies all five conditions. That is the point. The model tells you what absence of evolution would look like, so that a departure from it is evidence that one of the five is being violated, and the question is usually which.

The forces themselves divide neatly. Mutation makes new alleles but moves frequencies very slowly by itself. Gene flow makes populations more alike. Selection is the only one that is systematically related to fitness, and fitness means reproductive success and nothing else. Drift is the opposite of selection in a specific sense: it changes frequencies at random, its effects are strongest in small populations, and it can fix a harmful allele or lose a beneficial one because sampling does not care.

Speciation and phylogeny are the same story over longer time. Populations that stop exchanging genes diverge, and reproductive isolation is what makes the split permanent. A tree drawn from the result is read by shared ancestry, never by how close two names sit on the page.

What to hold

• Evolution is a change in allele frequencies in a population over generations, so an individual organism cannot evolve.
• Hardy-Weinberg assumes no mutation, no gene flow, no selection, random mating and a population large enough for drift to be negligible.
• Under Hardy-Weinberg, p + q = 1 for two alleles and the genotype frequencies are p squared for homozygous dominant, 2pq for heterozygous and q squared for homozygous recessive.
• The carrier frequency for a recessive condition is 2pq, not q squared: q squared is the frequency of affected individuals.
• A recessive disease affecting 1 in 10,000 gives q = 0.01, p = 0.99 and a carrier frequency of about 2%, which is roughly 1 in 50, so carriers vastly outnumber affected people.
• For an X-linked recessive allele, the frequency of affected males is q rather than q squared, because a male has only one X and cannot mask it.
• Non-random mating, including inbreeding, shifts genotype frequencies towards homozygotes but does not by itself change allele frequencies.
• Selection acts on phenotypes, so a recessive allele hidden in heterozygotes is largely invisible to it, which is why selection against a rare recessive is very slow.
• Heterozygote advantage maintains both alleles in a population, which is why an allele harmful in the homozygous state can stay common where it protects heterozygotes.
• Genetic drift changes allele frequencies by chance alone, has its largest effect in small populations, and can fix or eliminate an allele irrespective of its fitness.
• Bottlenecks and founder effects are drift acting through a small sample, so the surviving or founding group's allele frequencies need not resemble the original population's.
• On a phylogenetic tree, relatedness is read from the most recent common ancestor, and branches can be rotated at any node without changing what the tree says.
• Homologous structures reflect shared ancestry and divergent evolution; analogous structures reflect similar selection pressures acting on unrelated lineages.