Butchering science

This is just a mess.

I have written in the past about Jack Hudson’s tendency to butcher science. There are a lot of examples of him doing this, but one of the most egregious was when he concluded that because fruit fly populations under laboratory conditions come to allelic fixation at a different rate and/or way than asexual populations, that must mean there is some flaw in evolutionary theory. (In re-reading his post, I’m also seeing that he concluded something else equally egregious: He said that mutations which affect mRNA structure as opposed to protein sequence is evidence that random mutations cannot lead to new traits.) In short, it is highly evident that Jack read a popularized article about a recent study, glommed a few lines from the original research, and then went about drawing inept conclusions.

And now he’s back at it.

Let me give an actual summary of this most recent study first:

Stickleback fish are found all around the world. They exist in a number of streams, rivers, lakes, and oceans due to their great ability to adapt quite quickly to their environment. This opens up a great opportunity to take a look at their genes to see just what regions are evolving.

Now, what often happens in these sort of studies is that researchers will choose selected areas or candidate genes and compare them. It’s a tried and true method, but it probably isn’t the whole picture. While researchers can grow various species (usually of bacteria), objectively know how they’ve diverged and evolved under laboratory conditions, and then compare what they know to phenotypic changes brought on by alterations in protein-coding genes, there has been a push for a long time to sequence more and more full genomes. One result has been information overload (even when the full genome of something has not been sequenced; the technology that allows full genome sequencing also inherently allows easier partial sequencing), but that’s not a bad problem to have. So for this study, the biologists sequenced 21 three-spined stickleback genomes. Their goal was to determine the underlying molecular basis for adaptive evolution in the fish: Do they evolve by way of regulatory or coding changes?

What the team found was that 147 regions vary in freshwater versus marine stickleback populations. Of these 147 regions, 17% were linked to coding genes, 41% to regulatory regions, and 42% could not be classified cleanly (though, as the neat little graph under “Proportion of regulatory and coding change” in the paper says, they are probably regulatory).

So the big conclusion is this. Stickleback evolution is dominated by regulatory changes – changes involving areas which control genes. (Coding changes are still important, but this study indicates a possible shift in focus as it becomes cheaper and easier to sequence whole genomes.) The regions prevalent in stickleback evolution are relatively few; we keep seeing the same areas get tweaked over and over, leading to independent (and often convergent) solutions for the same sort of environments.

Now let’s look at Jack’s butchery:

But these findings are actually quite contrary to the sort of evolution often advocated by Darwinian evolutionists. Instead of incidental mutations coding sequences leading to the production of new proteins (and conceivably, novel structures and systems) the researchers found that the changes were primarily to the same sets of regulatory sequences in separate populations of sticklebacks…

While the researchers continue to use the term ‘evolutionary change’, the reality is this is nothing like the sort of change described by the modern evolutionary synthesis, a theory which relies on natural selection acting on genetic mutation.

To summarize this inanity: Jack is saying that evolution predicts that changes in species should occur almost exclusively by way of natural selection working on random mutation. He’s wrong. What evolution says is that change will occur by a number of mechanisms – random drift, hitchhiking via linkage, bottlenecking, horizontal gene transfer, and others. Random mutations culled by non-random selection will result in changes, certainly. And that’s what we see quite frequently in the laboratory and nearly 1/5 of the time in this stickleback study. However, the presence of other mechanisms is not somehow counter to evolutionary theory. Indeed, I think embryologists would be rather upset to learn that their field undermines evolution since the regulation of development – not necessarily or even usually by coding regions – has a huge impact on the way species change over time.

Yet I haven’t even gotten to the kicker:

The very fact that the researcher describes these as “key genes that control evolutionary change” contradicts the ordinary notion of evolution itself, which is purportedly an unguided process.

This reminds me Mary Midgley’s complete misunderstanding of Richard Dawkins’ The Selfish Gene. Midgley complained, chiefly, that Dawkins’ use of “selfish” was wrong because genes don’t have emotions. It was risible and I don’t know as Dawkins should have bothered responding. Naturally, I have to wonder if I should bother with the likes of Jack Hudson.

The language being employed by the scientists behind the stickleback study does not indicate that there is any guiding mechanism to evolution. They obviously are not claiming that regulatory genes direct evolution in a predetermined way. All they are saying is that these genes are a major factor in evolutionary change. It would be as if I said that my gene for lactase controls my tolerance for lactose (dairy products, more or less). That does not mean there is a little man sitting on a section of my DNA, tinkering away because he desires that I ingest milk.

Sorry to keep this going, but there are a lot of kickers in this one:

If natural selection acting on incidental mutations were actually capable of producing the radically different body plans, structures and systems we find throughout the plant and animal kingdoms, then we wouldn’t expect to see the consistent similarity of genetic modifications that we do with regard to the various populations of sticklebacks.

It would be generous to say this is a strain on logic. It absolutely does not follow that the predominance of regulatory genes excludes the importance of random mutation. Moreover, this study is not looking at millions of years, so extrapolation out that far should be constrained.

The changes wouldn’t be a matter of merely regulating extant genes, but the origination of new genetic capabilities.

Jack is, in essence, claiming that regulatory changes do not count as evolution. Unfortunately for him, we have thousands of different species of sticklebacks that attest to significant change over time – and now we know they’ve been doing it with a lot of help from their regulatory genes. So even if there was something to Jack’s claim on its own, it wholly crumbles when we hold it up to all the different stickleback populations around the globe.

Genetic sequencing continues to demonstrate that there are limits to biological variation.

This is in reference to specific creationist-proposed limitations, something not supported by an iota of this study. That is, this claim boils down to Jack saying that because sticklebacks evolve in a large number of ways by virtue of relatively few regulatory regions, species are constrained to microevolutionary changes. Again, this is a logic fail. The presence of changes by way of natural selection operating on regulatory genes does not exclude changes by way of natural selection operating on random mutation. Anyone who bothered to honestly look at this study would know that. (17% and perhaps more of the regions map to actual genes, for Christ’s sake.)

As I’ve said a few times before, what takes a creationist 30 seconds to say takes an educated person hours to untangle. And just as with my last post about butchered science, this didn’t take quite that long, but the sentiment remains true.