Liveblogging the conference: Jon Seger

Jon is a Utah biologist. His talk is on population genetics.

He is talking about an unusually clean evolutionary experiment that leaves natural populations just as messy as they were before. Real populations are so complicated they frustrate basic models and general principles of population genetics. The whale lice of right whales are the case study used. The speciated about 5 million years ago and have distinct habitats.[Coalescent theory has completely changed how we do this work.]

Okay: species are communities of genes, he says. Various problems - ring species, clines, yadda yadda. [Very pragmatic about species, as a good biologist should be.] Whale lice are crustaceans that form a white carpet on the surface of right whales. They have no free swimming stage so transmission is initially from mother to child. But some cross transfer from associates. There are three kinds of lice and three separate populations of right whales (Argentina, South Africa, Australia), and because we know the whale population and the number of lice per whale, we can a priori estimate the lice population. Almost no differentiation of lice between whales - they move so fast they are extremely well mixed. Drift is slow, as the population is large.

But across the equator, isolation has made their mtDNA different, dating to the closing of the Isthmus of Panama. About one million years ago a southern whale moved across the equator introducing the southern lice to the northern Pacific population. The mtDNA is reciprocally monophyletic for that group - not so the nuclear genes. They have a coalescence date way back in the past - tens of millions of years. The mtDNA is recently coalescent.

[I wish I could show you his figures.] Conclusions: marine lifestyles can "ablosh space"; given a large N drift is glacially slow; do they still belong to the "same" population if not gene flow for one million years? and genes and genomes coalesce at different rates; and probably no distinctive signatures in speciation events.

More like this