« Celebrating Science Seasonally | Main | This year in "intelligent design" »

This year in evolution, in Nature

The leading general-science journal, Nature, published many papers on evolution this year, two of which made it into their annual list of the editors' "favourites." One paper described laboratory experiments using the power of "directed evolution" to improve enzymes. The other explains the rapid evolution of new fish species in the wild. The editors also highlight two papers giving ice-core records of atmospheric carbon dioxide and methane over the last 800,000 years. These greenhouse gases both affect and are affected by life on earth, so they provide valuable context for recent evolutionary trends.

In a paper titled "Kemp elimination catalysts by computational enzyme design", Daniela Rothlisberger and colleagues started by designing enzymes using computer software. I don't know how the "RosettaMatch hashing algorithm" works, but the process seems to have a lot in common with natural selection's approach, selecting the best from "more than 100,000 possible realizations" rather than going directly from first principles to the ideal design. Furthermore, they noted that "our in silico design process seems to be drawn towards the same structural features as naturally occurring enzyme evolution."

Their next step was even closer to natural evolution. They started with their best design and generated lots of random mutants, testing each for activity. After seven rounds of this directed evolution, they got a 200-fold increase in enzyme activity. They noted that "in vitro evolution... is currently the most widely used and successful approach for refining biocatalysts."

Once again, nonrandom selection among random variants solves problems beyond the reach of (present) human ingenuity. How many more examples do we need before people stop saying "it's sophisticated, must have been designed" and instead say "it's sophisticated, must have evolved"?

In the second paper, Ole Seehausen and colleagues describe "Speciation through sensory drive in chichlid fish." For a species to split in two, interbreeding between two groups has to be low enough that they can evolve differences. This is easy if they are geographically separated, as on different islands or on continents drifting apart. But what if they're all fish in the same lake?

Even within a lake, there can be major differences in light conditions. Females often prefer brightly colored males, but the definition of "brightly colored" depends on lighting and on how sensitive female eyes are to different colors. They compared two related species, one with red males and one with blue males. These color differences in the males corresponded with differences in color perception by the females. The red fish were found at greater depth, where more red light penetrates.

Speciation depended on how rapidly light conditions changed with distance. If it didn't change at all, selection operated similarly across the range, giving no differences in light sensitivity. But apparently the same was true if light conditions changed too rapidly. In that case, fish had to cover a range of light conditions, so didn't evolve the specialization that would lead some females to prefer blue males and others to prefer red males.


A comment on light absorbtion in water - red light is absorbed at shallower depths, blues penetrate further. This can easily be seen when scuba diving - as one heads below 10m corals look yellow and blue, and bluer with increased depth. Red details on equipment look green, then blue. Illuminate them with a local source of light (a battery torch, say...) and reds will be seen again. So red fish may well be selected for at depth, but it cannot be because red light penetrates further, 'cause it doesn't. It's more likely that it's because reds are less visible at depth that red fish do well, but I'd need to think harder about why.

I can't read the original paper, only the abstract, so I don't know what the authors' comments were, but Dr Denison, you have either made or repeated an error above.

Showing that depth in the water column alone can provide allopatry is excellent, however!

Charlie, thanks for your interesting comment.

From the article: "The light climate of Lake Victoria is dominated by effects of particulate (nonphytoplankton) matter, selectively absorbing and scattering light of short wavelengths, causing successive shifts of ambient light towards longer wavelengths with increasing water depth"

Could your observations perhaps apply only to nonturbid water?

Adult stem cell research blew away those items like explanations of rapid fish species in 2008.

I agree that stem cells may turn out that to be of more practical value than information on fish evolution. Using evolutionary principles to design enzymes, on the other hand, has immediate practical value, while for stem cells we will have to wait and see.

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Type the characters you see in the picture above.