18 February 2008

Natura non facit saltum...umm?

The following is a guest post from Nicole Maturen, a PhD student in evolutionary developmental biology (evo-devo), who had to write a blog post as part of an application for an internship at an eminent open-access journal. Though publication of the post on an actual blog was not a requirement of the application, I thought it would be a real shame for such a gem to moulder away on a single lonesome hard drive. So, without further ado, over to Nicole! - nunatak

As noted on this blog earlier, this year’s celebrations of Darwin’s birthday were preceded by a bit of blog fanfare. I am happy to report that the goods were delivered during his Party at the Natural History Museum, London and that the warm-up toward 200 is going well.

Brian Charlesworth for PopGen opened the party-gritty with an overview of his field’s approach to the question of whether or not evolution makes jumps. The closest thing that he gave to an answer was the story of the Fisher --> Fisher + Kimura --> Fisher + Kimura + Orr geometric model of adaptation. QTL mapping data and this current model are in fairly close accord that the genetic change between (presumably closely-related) species comprises a small number of significant ‘hops’ followed by generous fine-tuning. So perhaps we can answer yes – or at least say that some steps that evolution takes are larger than others.

Brian also left us with two gems: 1) a careful disclosure of the assumptions behind the models that mold the way we think about these questions and 2) a handy review in case any of us had gotten a bit lost.

Next, David Stern for EvoDevo told several exciting stories (with data and pretty pictures). First, the story of Shavenbaby, a locus privileged among many in the genetic network of Drosophila by it’s switch-like position. It appears that Shavenbaby, which codes for a transcription factor controlling trichome development, has been turned on and off in the genus in a specific and reversible way multiple times. Upon closer inspection, we see that it actually consists of several switches; different parts of the gene which control distinct zones of its expression have been switched on and off separately. Thus, what appears from one vantage point to be a ‘monster’ mutation is actually the composite effect of several changes. It’s a question of scale, as we’ve heard twice before at Darwin’s birthday party, and as we read in this year’s assigned pre-party reading on the meaning of punctuated equilibria.

The mutated parts of the Shavenbaby gene are bits of sequence in the promoter region; it’s not the product of the gene, the protein molecule, which is different among these species, but rather the regulation of when and where the molecule is active. David also spoke about The Database of Evolutionarily Relevant Mutations (and welcomed any suggestions for a pithier name). One goal of assembling such a database was to address the question of how genes usually evolve new expression patterns – through mutations in regulatory elements, the favored hypothesis, or through mutations in the structure of a protein. One of the most interesting patterns to emerge from the data is a much higher incidence of regulation-type changes between species than within a species, as also found in this study.

In his rebuttal to Olivia Judson, Jerry Coyne argues that a jump, or macromutation, almost always produces a sickly organism (who is unlikely to pave the way for a new species) because of the multiple and necessarily deleterious side-effects of such a big-impact mutation. David Stern would agree, but he points out that all loci are not created equal, that is, they don’t occupy equivalent positions in the genetic network of an organism, nor do all mutations occur in equivalent positions in the genetic structure of a locus. Some changes, e.g. regulatory, in some types of genes, e.g. switches, produce clean(er), (more) functional mutants than average. David suggested that the mutations that lead to (macro)evolution might be those rare, non-pleiotropic ones, simply because they are non-pleiotropic (see this paper).

To conclude, David left us with a plea for MORE DATA about both the genetic architecture and the population dynamics of a range of organisms, to which everyone, even the spirit of Darwin himself, could happily raise a toast.


Richard Carter, FCD said...

Last month, John Wilkins over at Evolving Thoughts write an excellent post about the phrase natura non facit saltum, explaining how it doesn't mean what most people think it means.

colonix said...

wow. this is an incredible post, i really enjoyed this.