Phylogeny Friday - 27 April 2007

This week's phylogeny comes from this paper on molecular dating of speciation events. I won't be addressing molecular dating per se, but I will be dealing with what molecular clocks tell us. Like, do they actually reveal the speciation time of a pair of species?

The divergence date of a pair of species can refer to two things: when the two populations became two species (no longer exchanging alleles) or when the two genetic lineages split. The splitting of genetic lineages happens prior to the speciation event. That's because within a population there is variation throughout the genome. It's that variation, when it gets fixed between species, which contributes to the initial divergence between the species. Most of the molecular variation gets fixed randomly, but the variation originated prior to the split of the species. When we use divergence times to date speciation events, we're actually looking at the date when that variation arose.

i-2341dc55fac6efe27378f8de4f8068c5-species_gene_tree.jpg

This figure displays the aforementioned concepts. The red line shows the genetic relationships of members of Species 1 and Species 2, and the black outer lines show the relationships of the organisms. The molecular divergence of species 1 & 2 is marked by "Cc", while the actual point at which the lineages become true species is indicated by the blue line. Between Cc and the blue line, alleles are segregating in an ancestral population. Some of those alleles end up fixing along the lineage leading to Species 1, and some fix on the lineage leading to Species 2.

For recently diverged species, this can be problematic. Take, for example, the hypothetical species pair in the figure. The amount of time from the present back to the speciation event is approximately equal to the amount of time from the speciation event to the molecular common ancestor (Cc). That means that estimates the divergence time of the two species using molecular data will not be a good approximation of the speciation time of the two species; the molecular divergence time of the species will be a gross overestimate of the speciation time.

A good example of the disagreement between speciation times and divergence times is the human-chimpanzee species pair. Estimates of the human-chimp divergence using molecular data center around 6 million years ago (see here for an example). Recently, a paper was published that dated the speciation time of humans and chimps at approximately 4mya (available here). These are not conflicting estimates, as the divergence time is expected to predate the speciation time.

If you followed the discussion this far, it should be clear to you that this problem goes away when we're dealing with divergence times further back in the past. The time between Cc and the speciation event stays constant, but the distance from the present to the speciation event increases over time. Eventually, the time from the present to the speciation event dwarfs the time from Cc to speciation, making the estimate of divergence time equivalent to the estimates of speciation time.


Hobolth A, Christensen OF, Mailund T, Schierup MH. 2007 Genomic relationships and speciation times of human, chimpanzee, and gorilla inferred from a coalescent hidden markov model. PLoS Genet 3: e7 doi: 10.1371/journal.pgen.0030007

Kumar S, Filipski A, Swarna V, Walker A, and Hedges SB. 2005. Placing confidence limits on the molecular age of the human-chimpanzee divergence. PNAS 102: 18842-18847. doi: 10.1073/pnas.0509585102

Pulquerioa MJF and Nicholsa RA. 2007. Dates from the molecular clock: how wrong can we be? Trends Ecol Evol 22: 180-184. doi: 10.1016/j.tree.2006.11.013

More like this