Back in the middle of last month, I had a few things to say about Casey Luskin (DI flak) and his understanding of so-called junk DNA. It's now the middle of the month again, and Casey is again talking a lot - and understanding very little - about "junk" DNA. Larry Moran has a post up where he tries to educate Casey about the fact that a hell of a lot of DNA is still, at least as far as we know, junk. I'm going to take a look at something a little bit different - one of the methods scientists use to identify areas of "junk" DNA that have important functions. It's a pretty cool way of doing things, but it's not one that Casey likes to talk about - because it's really one of the finest examples of how our understanding of Darwinian evolution has lead to new discoveries about living things.
First, I should probably take a minute or so to talk about what Casey means when he uses the phrase "junk" DNA. In Casey's world, there are two classes of DNA: coding regions, and "junk". That because, a long time ago, scientists thought that the only important DNA was the DNA in coding regions (that's the DNA that's used to make proteins) and that everything else was unimportant. Someone used the phrase "evolutionary junk" to describe the non-coding regions, and Casey (and other anti-evolutionists) somehow or another decided that the "neo-Darwinian paradigm" demanded that non-coding DNA was junk, and interfered with research that might indicate otherwise.
That's the story in the world of the Discovery Institute. In Realityworldland, things are just a tad bit different.
To begin with, scientists began to realize that at least some non-coding regions had important regulatory functions fairly early on - without them, the DNA can't be used to make proteins. In other cases, stretches of DNA seem to function as spacers. In these cases, the DNA sequence doesn't matter much (if at all). Instead, the important thing is that the spacer ensures that, for example, a regulatory region sits the necessary distance away from the coding region of the gene. Casey lumps all of those types of DNA together with DNA that we don't know the function of and DNA that we are pretty sure has no function, labels all of it junk, and claims victory for Intelligent Design anytime any of it turns out to do something important. That's more or less the level of integrity that we've come to expect from Casey, of course, but it's still somewhat disappointing that he can't muster the strength of character to argue honestly.
But, as Arlo said, that's not what I came to tell you about.
Came to talk about the junk - and how to tell when it really isn't.
There are some stretches of DNA that do something important, but we don't know what. Scientists are working on figuring it out, but right now all we know is that they're important. We know this because we understand how evolution works, how natural selection works, and can - do - use that understanding to investigate DNA.
Here's how:
We know that mutations happen. DNA does not copy itself perfectly, so segments of DNA tend to change over time within populations, unless natural selection acts to prevent that. When a segment DNA does something that's very important - so important that the organism can't reproduce well without it - that segment of DNA doesn't evolve much. Exactly how much it changes depends on just how important its function is, and on just how much change the segment can tolerate while still working well enough.
We are now able to sequence not just small stretches of DNA, but the entire genome of an organism. It's not easy yet, and usually involves a lot of scientists working very hard for a long time, but we can - do - do it. We've sequenced the entire human genome, the entire chimp genome, a rat, a mouse, chicken, -the list goes on, and keeps growing. Last week, the genome for an anemone was published.
We are able to compare the genomes that are sequenced. This is also a bit challenging, and can require a lot of computer power and a lot of time, but it's done routinely at this point. We can take newly-sequenced genomes, compare them to other sequenced genomes, and look at the similarities and differences. In some cases, scientists have discovered that some non-coding regions are very, very similar between species that are very different. That can mean only one thing: those non-coding regions must be doing something important.
There is simply no other possibility. Natural selection is as much a force of nature as gravity, and if selection was not acting to "purify" those regions - to keep them unchanged - they would not be so similar between different species. Knowing this, scientists can begin to work to discover the functions for these areas.
In some cases, our understanding of evolutionary relationships can help narrow down the possible functions for the DNA in question. For example, there are regions of DNA with unknown functions that are very similar in mice and men, chimps and chickens, dogs and dogfish - but are not known at all in flies and worms. We don't know just what they do, but we can safely predict that it's something that's important to vertebrates and irrelevant to inverts.
That's how evolutionary science drives the discovery of functions for some of the so-called "junk" DNA. I'm still waiting to hear what Intelligent Design could add to that. I'll be waiting for a while, I expect.
- Log in to post comments
Luskin would seem to have it that nobody ever thought that noncoding DNA might have a function because evolutionary theory predicts that it is all junk.
He is wrong on both counts. Evolutionary theory predicts that there will be some junk (in the original sense of DNA that really has no current function whatsoever), but it doesn't offer a clear prediction as to what fraction of noncoding DNA will have function and what fraction will not. It's been known since the very earliest days of molecular biology, going back to early investigations into bacterial genetics, that some noncoding DNA has important regulatory functions.
And I've never know any molecular biologist to assume that noncoding, or even intronic, DNA was necessarily without function. Everyone I've ever known who worked in the area was clearly aware that there could be regulatory regions or other important code hidden in introns, but nobody know how to recognize them--it was a problem awaiting an experimental solution.
The most frustrating aspect of this debate is that IDists insist they can claim ID predicts little to no "junk" DNA in the first place even though they claim that they cannot step into a designer's shoes and predict what a designed lifeform should look like.
Why shouldn't DNA have any junk in it, even if they're correct and some cosmic entity created it?
Some IDists claim that front-loading could have been used to embed all the necessary functions required for the "evolution" of life over the past 4 billion years, so why couldn't "junk" DNA regions simply be empty, left-over buffers that used to contain front-loaded instructions which are no longer required, and have since been disused leaving only the random junk of left-over processing?
It's all nonsense of course, but it shows how easily this junk DNA "prediction" falls flat on its face. Junk DNA cannot falsify ID, hence it is useless as a prediction.
IDists like to argue that a design would show evidence of efficiency, which "junk" DNA would contradict, but since when is a 4 billion-year-old process that gets us from "goo to zoo to you" (to use a favorite creationist expression) the model of efficiency? Human designers may be slow sometimes, but four billion years? Come on.
No, no--argue the IDists--ID can only detect design, not how the designer goes about his business, there may be good reasons for taking four billion years. Exactly, we cry. Now, what were you saying about all that junk DNA again...?
Have you considered expanding this into a Basic Concepts post? Because it explains the concept beautifully.
You can believe anything you want at Casey's Restaurant.
You can believe anything you want at Casey's Restaurant.
Walk right in if you're a hick, and believing in ghost's is your kind of trick.
You can believe in anything you want, at Casey's Restaurant.
BTW - Are Casey and officer Obie related?
Note however, that the evolutionary conservation of sequence only implies that this sequence ***was*** important ***on evolutionary timescales***. This sequence may have no functional consequence whatsoever on timescales of the individual animal. For humanity at large over millions of years, it might matter. For you and me personally, it can still be junk. Additionally, it may have no evolutionary importance either today as the selective forces that conserved it may no longer exist.
In some cases, scientists have discovered that some non-coding regions are very, very similar between species that are very different. That can mean only one thing: those non-coding regions must be doing something important.
There is simply no other possibility. Natural selection is as much a force of nature as gravity, and if selection was not acting to "purify" those regions - to keep them unchanged - they would not be so similar between different species.
How does the observation that deleting a few highly conserved non-coding regions causes no adverse phenotypic affects on the organism?
http://www.ncbi.nlm.nih.gov/sites/entrez?db=pubmed&cmd=Retrieve&list_ui…
Reading the paper, they don't say what mice they used, but glancing at this page* tells me the average genome size for mice is around 3.0 pg, converting into basepairs, that's about 2934000000 basepairs. The reserachers deleted two regions that totaled to 2356000 basepairs, so they deleted around 0.080% of the mouse's genome.
*http://genomesize.com/results.php?page=1
If these regions are serving important functions (ie being maintained by stabilizing selection), why do they appear extraneous?
Also, much of what Luskin claims as new research making us realize our mistakes is the outcome of the human genome project. We weren't doing this research 20 years ago because we thought the DNA was junk. We were studying non-coding DNA avidly, I've shown Mims at the very least how to find this information.
The reason we can do these comparisons across many individuals in the non-coding regions is the human (and other animal) genome projects have given us so much information about conservation and a basic template to compare things to. This research isn't coming out late because we were sitting on our hands ignoring non-coding DNA. It's coming out because we have the genome!
What the IDiots are forgetting, is that without the idea of common descent, there would be no reason to compare genomes of different organisms. If each and every species was uniquely designed, what would be the point?
If the IDiots were in control, it would probably be a crime to compare genomes. Who knows what heritical thoughts might creep in if people were actually able to look at what actual genomes say about what organisms descended from.