How Predictable Is Evolution?

Evolutionary biologists like to say that mutations are random but that selection is not; species are crafted by their environments. But if this is the case, why is it so hard to predict evolution? A recent genetic analysis of stick insects provides an important clue.

At the heart of this issue a phenomenon known as parallel evolution — when similar traits evolve and emerge independently in related populations (as opposed to convergent evolution, which involves the emergence of similar traits in unrelated lineages, or evolutionary relay, in which independent species acquire similar characteristics through their evolution in similar ecosystems, but not at the same time (e.g. dorsal fins of sharks and ichthyosaurs)).

A related aspect of parallel evolution is parallel speciation. This is when related populations experience changes in their physical appearance and psychological characteristics despite the fact that they're being exposed to the same environmental conditions.

But evolutionary biologists have not been able to figure out the degree to which the same or different sets of genes control these parallel morphological changes. By solving this mystery, biologists could potentially determine the repeatability of evolution and whether or not organisms would evolve in the same way if we were able to rewind the clock of evolutionary history.

To learn more, Patrik Nosil, an evolutionary biologist at the University of Sheffield in the United Kingdom, sequenced the genomes of stick insect "ecotypes" that emerged via parallel speciation.

From the University of Notre Dame's release:

...Feder and colleagues transplanted different "ecotypes" of Timema specialized to feed on two different species of plants between the plants at a field site in southern California. Through Next Generation DNA sequencing of the offspring of surviving individuals, they measured the degree to which genes across the entire genome of the stick insects changed in response to the experimental manipulation. These results provided a base line to compare to patterns of genomic divergence observed in nature among a series of geographically separated pairs of Timema populations attacking the two plants.

The researchers reported finding that early stages of parallel speciation in the stick insect involved mostly non-parallel genetic divergence between the ecotypes. However, they also detected parallel genomic divergence for a subset of genes across populations involving mostly changes in protein coding regions of genes having specific molecular functions.

So, some genetic changes are shared, while others differ when populations evolve similar adaptive traits in parallel. Specifically, the researchers found that only 17% of the stick insect DNA had changed in the same way, suggesting that, while some evolution in the genes leading to host specialization is predictable, the vast majority of the changes are random.

In other words, the predictability of evolution is exceedingly low. Because environments offer a myriad number of adaptive opportunities, the subsequent space of all possible viable mutations (and thus phenotypes) is absolutely huge.

Read the entire study at Science: "Stick Insect Genomes Reveal Natural Selection's Role in Parallel Speciation." Supplementary info via ScienceNews.

Image: Acanthoxyla prasina or prickly stick insect, found in Fairfield, Otago, New Zealand, about 10 June, 2012. Photo by Alan Gilchrist.