What Are The Chances? Mathematician Solves Evolutionary Mystery

NewsWise ^ | 26 September 2003 | Staff

[Sabertooth]

We don't trouble ourselves that H2O is liquid only over a limited range of temperatures, but some get upset over the concept that evolutionary clocks can only be used to fill in gaps where other dating methods have supplied the range.

Doesn't upset me.

However, when I see unqualified claims made on the basis of evolutionary clock computations, then I am seeing sloppy writing and/or thinking that is ostensibly for the furtherance of science, but is actually a hindrance to clear scientific thinking.

That kind of garbage needs to be rooted out.

[<1/1,000,000th%]

Not a very informative article.

Where's my Cheetos?

[Doctor Stochastic]

There are many articles recently about calibration of various molecular (or evolutionary) clocks. A search of PubMed turns up dozens in the last few years.

[js1138]

There's lots of bad science writing in the general press. If you want a general source of accessable science writing I'd recommend sciencenews.org and their magazine. They make mistakes and have occasional sloppy writing, but their readers call them on it, and they print the letters.

[patriot_wes]

"To a certain extent, the answer lies in simple probability, says Pinelis. Say you have two species of fish swimming in a pond, the carp and the perch, and it might be equally likely that one of them will evolve a third species. Say the goldfish evolves from the common carp, and suddenly you have three fish species in your pond."

SOLVED....the evolution of democRAT presidential cantidates

[RadioAstronomer]

Nice read. :-) Thanks for the ping.

[jennyp]

Sounds like some common sense math. It's a double-edged sword, however, for another evolutionary notion currently in vogue, that of evolutionary clocks.

Just as we don't accrue wealth at constant rates, living fossils demonstrate that species don't accumulate mutations at constant rates, genera don't accumulate species at constant rates, etc. Evolutionary clocks are not analogous to radiometric clocks. The observed fact of relative evolutionary stasis vs. that of evolutionary dynamism calls conclusions based on evolutionary clocks into doubt

Ah, but if you had been a regular reader of, ahem, Creation/Evolution: The Eternal Debate you'duv seen the news of the new study of fast-mutating genes, where they claim that fast-mutating genes actually create more accurate deep-time molecular clock readings than slow-mutating genes:

However, Hilu and his colleagues have come up with a new approach using rapidly evolving genes to understand deep-level relationships. Those genes mutate at higher rates than the slowly evolving genes. Although evolutionary biologists previously thought rapidly evolving genes would give a misleading picture of deep evolutionary history and were useful only in more recent evolutionary events such as evolution at the species and genus levels, Hilu has demonstrated that as few as 1,200 nuclear-type bases of a rapidly evolving gene such as matK, a gene in the chloroplasty genome, will give a tree of angiosperm that is far more robust than that obtained from 13,400 bases of several slowly evolving genes combined.

With this new approach, Hilu said, scientists will be able to sample many more species, and the process will be much more economical. "This does not mean slowly evolving genes are useless," Hilu said, "but a combination of the two could give us information at different evolutionary levels."

Hilu has found that the quality of the signal is better in rapidly evolving genes due to tendencies towards neutrality and lack of as many strong functional constraints as in slowly evolving genes. He also found that rapidly evolving genes provide more characters because they keep mutating more quickly. "Between the quality and the quantity, we were able to obtain more historical signals from rapidly evolving genes," he said.

If this finding holds up, it should resurrect the notion of the molecular clock as an accurate tool for discovering phylogenetic relationships, even ones that diverged hundreds of millions of years ago or more.

[BiffWondercat]

Thanks for the ping! The fish species of Lake Victoria in Africa as well as the Amazon show how this can apply in the fish example. But y'all probably knew that already....

[colorado tanker]

Huh. And I thought the best example of the "living-fossil" phenomenon was Jimmy Carter.

[Yeti]

it's highly likely that any individual walk with be biased in one direction or the other. The totalitiy of walks is symmetric, however.

Rosencranz and Guildenstern have evolved.

Some imbalance seems intuitively right to me, if for no other reason than that some mutations might, by their nature, expose a species to environmental factors which will lead to more mutation, or expose different populations to diverse culling mechanisms. Other mutations might lead a species to an environment where there is less exposure to mutagenic phenomena, or where most mutations will be very disadvantageous, and quickly culled.

But there are many other reasons, and a perfect balance everywhere would be the odder occurence, in a system this large. Never getting heads twice in row would be very surprising in a series of a billion coin-flips.

[sleavelessinseattle]

Like, say, some family trees in Arkansas that resemble telephone poles?

Don't ya mean Hula Hoops?

[donh]

hmm. I wouldn't say the case is made here. Some genomes are going to have better meta-evolutionary strategies then others, and will prosper by prospecting for new niches to fill. More narrowly focused genomes will prosper best by honing themselves for being better at occupying the niche they are already in. Critters ain't necessarily markov chains at their fundament, just because a markov model seems to accurately reflect their behavior.

[donh]

Evolutionary clocks are not analogous to radiometric clocks

The headline news of the evolutionary clock relates to the relative ordering of species in the tree of life, not the exact chronology of species.

[PatrickHenry]

Farewell, old thread. Final bump.