So Much Junk in the Genome... and the Press
Science journalism is a business filled with a few bright shining stars standing amidst a lot of writers whose stars are... let's just say, they don't shine as bright. Concerning the latter, there is a recent article in Wired magazine titled, One Scientist's Junk Is a Creationist's Treasure. It's your standard attempt at journalistic "balance" that puts crackpots on an equal playing field with actual scientists whose work the former group misrepresents. There's no excuse for this when a little background knowledge and a little more attention to what the real scientists are saying should show why the creationists are spouting nonsense.
As the title should tell you, the article has to do with so-called "junk DNA" and a recent paper concerning the opossum (Monodelphus domestica) genome. The authors of the paper found that a small fraction of transposable elements shared by the opossum and human genomes appear to contribute to host fitness, apparently by contributing to gene regulation. (Update: That particular paper isn't involved with the opossum genome project; the conserved sequences are found within "boreoeutherians" which include primates, rodents, and carnivores -- not the opossum.) This is a highly interesting if not exactly Earth shattering find.
Unfortunately, every time a new study comes along showing that some small fraction of so-called "junk DNA" turns out the have a function, the ID people do a strange sort of victory dance, as if this somehow proves that they've been right all along. In fact this is starting to become a frequent talking point with them. As with most creationists myths it's taken on a life of its own, and I'm sure we'll see it wandering around like a zombie for years to come in spite of the fact that it was DOA from the get-go. The new paper of course doesn't support ID by any stretch of the imagination, nor do any recent findings concerning junk DNA, but the author of the Wired piece, Catherine Shaffer, just credulously repeats claims made by Michael Behe and Stephen Meyer as if they had some measure of legitimacy. Which is why it's got Paul Nelson crowing about it. Below I will do the work that Shaffer didn't and explain just how wrong these guys are.
The problem with the argument being put forth by the ID people is as follows:
1. The whole history behind the "junk DNA" concept will have to wait for another day, but suffice it to say that the standard ID line on this is completely untrue. They claim that "Darwinists" thought the genome would be mostly non-functional, and that the "Darwinists" were surprised, recalcitrant, and stubborn to face the facts when things turned out differently. The article quotes Michael Behe as saying, "From the very beginning Darwinism thought whatever it didn't understand must be simple, must be nonfunctional. It's only in retrospect that Darwinists try to fit that into their theory."
As usual, Behe has his facts wrong. Prior to the advent of genomics, most strict Darwinians (i.e. those who believe that natural selection is everything) thought that the genome would be highly efficient and streamlined, that selection would mercilessly weed out any useless or wasteful sequences. Indeed, that largely seemed to be the case with bacteria. Although the strict Darwinian viewpoint had been losing ground by the early 70s, it was still quite puzzling when around 95% or more of eukaryotic genomes turned out to be non-coding. It required the non-Darwinian idea of neutral theory and much later the idea of selfish DNA to make sense of it all. To the extent that evolutionary theory had to retrospectively account for the evidence, Behe has it exactly backwards. The predominant adaptationist perspective had to give way to one that allowed for neutrality, contingency, ecology, and structural constraints. Only then did it occur to biologists that the whole genome need not be functional. The ID argument here is not only based on a false premise, it's an almost exact inversion of the truth.
2. Even more absurd, the article quotes Stephen Meyer as saying that, "It [this recent research] is a confirmation of a natural empirical prediction or expectation of the theory of intelligent design, and it disconfirms the neo-Darwinian hypothesis."
This, too, is utter nonsense. There is no logical connection between ID "theory" and the idea that the genome must be devoid of non-functioning DNA, because according to the IDists themselves, the "designer", along with its methods and its motives, cannot be defined. Nor have they come up with any model, coherent or otherwise, for how it was that living organisms came to be in their present state. Therefore, there aren't any constraints on what the "designer" may or may not do, and it (or they) could just as easily have added a bunch of non-functional DNA as made a genome that was 100% functional. Indeed, the IDists invoke this excuse all the time when dealing with the fact that many things in biology (not least of which is the genome) appear to be poorly or haphazardly designed. They even go so far as to say that it is a theological belief to state that a designer would have made efficient, highly functional designs. (One wonders how they could know it's theological if they don't know who the designer is.) We're not supposed to presume that the designer does things that make sense to us mere mortals. You see, he works in mysterious ways.
So Meyer's "prediction" here isn't a prediction at all, it's entirely post hoc reasoning. That makes Behe's quote from above not only wrong, but also deliciously ironic. We knew long before the ID movement began that at least some non-coding DNA would turn out to be functional, and molecular biologists were already busy searching for those functions back when the IDists were still calling themselves creationists. And now the IDists come along and retrospectively claim that they were the ones who predicted that non-coding DNA would be functional. Talk about shameless.
As for the bit about the research "disconfirm[ing] the neo-Darwinian hypothesis", that's so wildly wrong it doesn't require a detailed rebuttal. There's nothing about the neo-Darwinian synthesis that requires the genome to contain large amounts of non-functioning sequences. As previously stated, the default assumption among selectionists was that the genome would be fully functional. Meyer is simply engaging in dishonest and ignorant hyperbole.
3. Even if we assume for some reason that ID really does predict that all of the genome is functional, then so much the worse for ID. Neither this study nor any of the others that have appeared in recent years demonstrate that the entire genome is functional. What they show is that only a tiny percentage of the genome is functional. It's just a slightly larger tiny percentage than before. As the authors of the present study note, a mere 5% of the mammalian genome is conserved, suggestive of a function. Only 2% is protein coding, so the functional bits they're finding merely account for some of the discrepancy between the 2% that code for proteins and the 5% that we expect to be functional. I doubt there exists anywhere a competent molecular biologist who believes that all DNA is functional. The evidence strongly shows otherwise.
Not only does the vast majority of the genome have no known function, we have good reason to believe that much of it won't have a function because it consists of broken viruses, elements that rapidly duplicate, or degenerate gene copies. Many of these sequences are getting duplicated and deleted all the time, and vary not only between closely related species, but even between individuals within a species. Yet losing these sequences appears to have no consequence, and gaining new ones often causes disease by interrupting other sequences that are useful. Additionally, many non-coding sequences, particularly those from degenerate viruses and pseudogenes, accumulate mutations at the neutral rate. This wouldn't be happening if they were functional. Contrary to what the IDists believe, the non-functionality of these sequences isn't assumed by default, it's based on what we know about them.
4. What makes all this exceptionally ridiculous is that the present study has inferred the functionality of some non-coding DNA based on its evolutionary conservation. In other words, if humans and opossums have stretches of DNA that are highly similar in sequence, then we conclude that selection must have been preserving those sequences during the eons since humans and opossums split off from a common ancestor. (Update: Again, the Lowe et al. paper did not include the opossum genome, but rather used the dog genome as the most distant from humans. Just think "dog" instead of "opossum".) Given that Meyer and Nelson reject common ancestry, this particular method of analysis isn't supposed to work according to their belief system. As Gill Bejerano, one of the study's coauthors put it, "If you disbelieve this process, then from your perspective, we haven't found anything interesting in the genome." This should have tipped off the Wired author that she was dealing with crackpots, but for some reason it didn't.
___________________________________________________________________________
So in conclusion, the IDists are claiming that they made a "natural empirical prediction" that has been successfully confirmed. The fact is however that they didn't predict it, it isn't correct, it doesn't contradict mainstream theory, and the supposed confirmation is meaningless according to their own belief system. If that's not enough to convince you these guys are nothing more than rank propagandists, not to worry: Stephen Meyer is working on a whole book of this nonsense called The DNA Enigma. We'll get to see some junk for sure.
P.S. Larry Moran has more about the Wired article. And T. Ryan Gregory gives a brief history of junk DNA, again dispelling the IDist myth that "Darwinists" thought it was all non-functional. See also his onion test.
42 Comments
SteveF · 14 June 2007
As I mentioned at Larry's blog, this new paper is significant:
http://www.nature.com/nature/journal/v447/n7146/abs/nature05874.html
We report the generation and analysis of functional data from multiple, diverse experiments performed on a targeted 1% of the human genome as part of the pilot phase of the ENCODE Project. These data have been further integrated and augmented by a number of evolutionary and computational analyses. Together, our results advance the collective knowledge about human genome function in several major areas. First, our studies provide convincing evidence that the genome is pervasively transcribed, such that the majority of its bases can be found in primary transcripts, including non-protein-coding transcripts, and those that extensively overlap one another. Second, systematic examination of transcriptional regulation has yielded new understanding about transcription start sites, including their relationship to specific regulatory sequences and features of chromatin accessibility and histone modification. Third, a more sophisticated view of chromatin structure has emerged, including its inter-relationship with DNA replication and transcriptional regulation. Finally, integration of these new sources of information, in particular with respect to mammalian evolution based on inter- and intra-species sequence comparisons, has yielded new mechanistic and evolutionary insights concerning the functional landscape of the human genome. Together, these studies are defining a path for pursuit of a more comprehensive characterization of human genome function.
From a BBC article:
"A close-up view of the human genome has revealed its innermost workings to be far more complex than first thought.
The study, which was carried out on just 1% of our DNA code, challenges the view that genes are the main players in driving our biochemistry.
Instead, it suggests genes, so called junk DNA and other elements, together weave an intricate control network.
The work, published in the journals Nature and Genome Research, is to be scaled up to the rest of the genome."
http://news.bbc.co.uk/1/hi/sci/tech/6749213.stm
Randall · 14 June 2007
To be fair, just because a DNA sequence mutates at a neutral rate doesn't mean it's non-functional. It could be very important that control sequence A and coding region B be 1000-1100 base pairs apart; which base pairs doesn't matter, so that region mutates freely.
Of course, that doesn't account for repeated motifs, obvious virus lysogenic-like material, pseudogenes, etc. I'm just pointing out that mutation doesn't necessarily change "functional" DNA.
Hawks · 14 June 2007
Given that Meyer and Nelson reject common ancestry, this particular method of analysis isn't supposed to work according to their belief system.
And given that ID, as you said, does not say anything about the designer and so says nothing about whether or not common ancestry is "true", no one is justified in using this study to claim that ID predicted anything. You can make some assumptions about the designer and accept that common ancestry has in fact happened (as Behe seems to have done), but then you are just justified in saying that "Beheism" predicts it.
ID predicts NOTHING!!!
Bond; James Bond · 14 June 2007
I believe this recent study is very interesting to your topic in that it indicates that 100% of the genome may be poly-functional.
Would 100% poly-functionality in the genome be a evolutionary prediction or a ID prediction?
Encyclopedia Of DNA: New Findings Challenge Established Views On Human Genome
The new data indicate the genome contains very little unused sequences and, in fact, is a complex, interwoven network. In this network, genes are just one of many types of DNA sequences that have a functional impact. "Our perspective of transcription and genes may have to evolve," the researchers state in their Nature paper, noting the network model of the genome "poses some interesting mechanistic questions" that have yet to be answered.
Other surprises in the ENCODE data have major implications for our understanding of the evolution of genomes, particularly mammalian genomes. Until recently, researchers had thought that most of the DNA sequences important for biological function would be in areas of the genome most subject to evolutionary constraint --- that is, most likely to be conserved as species evolve. However, the ENCODE effort found about half of functional elements in the human genome do not appear to have been obviously constrained during evolution, at least when examined by current methods used by computational biologists.
According to ENCODE researchers, this lack of evolutionary constraint may indicate that many species' genomes contain a pool of functional elements, including RNA transcripts, that provide no specific benefits in terms of survival or reproduction.
The ENCODE consortium's major findings include the discovery that the majority of DNA in the human genome is transcribed into functional molecules, called RNA, and that these transcripts extensively overlap one another. This broad pattern of transcription challenges the long-standing view that the human genome consists of a relatively small set of discrete genes, along with a vast amount of so-called junk DNA that is not biologically active.
http://www.sciencedaily.com/releases/2007/06/070613131932.htm
Steve Reuland · 14 June 2007
Jerry · 14 June 2007
Here's a relevant paper that deserves more attention:
Marcelo A. Nóbrega, Yiwen Zhu, Ingrid Plajzer-Frick, Veena Afzal and Edward M. Rubin (1994). Megabase deletions of gene deserts result in viable mice. Nature 431, 988-993.
Mike Klymkowsky · 14 June 2007
Jerry has it right, I believe -- that genomic sequences are transcribed does not prove that they are functional. Only carefully done mutational/deletion studies can do that.
Believe it or not, though, scientists (and their university/institutional PR departments) like to make prosaic observations into earth shattering events.
At the same time, I cannot see how this relates to any prediction may by IDiots - or am I missing something deep here?
Nick (Matzke) · 14 June 2007
Marc Geerlings · 14 June 2007
> At the same time, I cannot see how this relates to any prediction may by
> IDiots or am I missing something deep here?
Yes, you miss something deep here. Like vestigal organs, creationists/IDers oppose every claim that something in the (especially human) body is not perfect created. Surely a perfect creator wouldn't deal with 'junk'
I think they see this as a prediction by them and as a consequences every hint of something considered as junk or vestigal by those godless scientist turning out functional or with a purpose is a victory for them. They predict that everything will come out as perfect functional the end.
Anna Lytickle · 15 June 2007
SteveF · 15 June 2007
Nick,
According to Steve (and a linky in the article), the relevant paper is this one on the opossum:
"Thousands of human mobile element fragments undergo strong purifying selection near developmental genes."
Whereas I referenced a paper that came out yesterday in Nature (and this is what the BBC article concerns itself with):
"Identification and analysis of functional elements in 1% of the human genome by the ENCODE pilot project."
I agree with Steve that the Wired article is feeble, but I was wondering what people thought about this additional study.
SteveF · 15 June 2007
This is taking balance to the extreme (i.e. completely making stuff up). Shaffer says in the comments:
"I interviewed five scientists for this article. Dr. Francis Collins, Dr. Michael Behe, Dr. Steve Meyers, Dr. T. Ryan Gregory, and Dr. Gill Bejerano. Each one is a gentleman and a credentialed expert either in biology or genetics."
Nick (Matzke) · 15 June 2007
Oops, my bad -- I was somehow confusing different things. The marsupial genome paper actually came out long ago (well, early May) so I tuned it out for some reason. I agree the Wired article is about the marsupial genome -- I guess we can add that it is six weeks late in addition to everything else.
Anyway, people have been arguing about the press coverage of the ENCODE project on various blogs -- no one has a problem with the paper, which is interesting but not any kind of revolution in terms of understanding junk DNA.
The basic point of the ENCODE paper as far as I am interested in it:
1. They looked at a somewhat-but-not-completely representative 1% of the genome.
2. Most of the sample DNA is transcribed to RNA
2a. However I am told that much of that is low frequency transcription and could easily just be random due to the generality of the relevant enzymes
3. Only about 5% of the sampled DNA is evolutionarily conserved according to the ENCODE study (ah -- the "present study" thing in the present PT post is what got me mixed up), which does not help arguments which say that all DNA is functional.
For more see genomicron: http://genomicron.blogspot.com/2007/06/more-about-encode-from-scientific.html
SteveF · 15 June 2007
Thanks Nick. There is also an article on it in today's Science. Some people are predicting revolution, others being more cautious:
Given the traditional gene-centric perspective, that finding "is going to be very disturbing to some people," says John Greally, a molecular biologist at Albert Einstein College of Medicine in New York City. On the other hand, says Francis Collins, director of the National Human Genome Research Institute (NHGRI) in Bethesda, Maryland, "we're beginning to understand the ground rules by which the genome functions."
http://www.sciencemag.org/cgi/content/full/316/5831/1556
(subscription required I think)
Popper's ghost · 15 June 2007
Bond, James Bond · 15 June 2007
Steve Reuland wrote:
It indicates nothing of the sort. If you read this article in Nature News for example, you learn that even the most radical of ENCODE researchers think that up to 20% of the genome may be functional. The rest think it's lower.
If ID wants to claim 100% functionality of the genome as a prediction, fine with me. We can then safely conclude that it's wrong.
Yet the first sentence in the blog entry states:
The new data indicates the genome contains very little unused sequences and, in fact, is a complex, interwoven network.
I don't think "very little unused sequences" means 20% functionality as you imply! I think very little unused sequences looks very suspiciously like ther fully functional genome predicted by ID.
Nick Matzke asks?
What the heck does "100% poly-functionality" even mean?
Yet, The first sentence in the second paragraph of the blog entry states:
The ENCODE consortium's major findings include the discovery that the majority of DNA in the human genome is transcribed into functional molecules, called RNA, and that these transcripts extensively overlap one another.
"EXTENSIVE OVERLAPPING" is what poly-functionality means!
whats more is that ID will expect the data compression of the genome to be found to be several codes thick for much of the genome!
Andrea Bottaro · 15 June 2007
Steve Reuland · 15 June 2007
ck1 · 15 June 2007
The article in Nature on the massive ENCODE project is an an overview.
This overview is linked to a series of focused articles dealing with specific aspects of the project - these articles have now been published in a dedicated issue of Genome Research:
http://www.genome.org/current.shtml
The individual articles are open access.
Heavy-duty stuff and a lot to digest, but the issue begins with a "perspective" essay that summarizes the key findings and puts them into context. There are also two commentaries - one on what these findings mean for the definition of "gene" and the other is about phenotypes.
harold · 15 June 2007
James Bond -
You seem to be missing a major point here.
Non-coding DNA is one of many massive problems for ID/creationism. Why would a designer use it every time he magically created a eukaryotic cell, if it was in any way imperfectly functional?
It is NOT a problem for science, nor is its lack - prokaryotes don't have it.
When creationists gloat over some suggestion of function in non-coding DNA, they're not scoring a point, they're merely trying to get one major (and valid) penalty against them overturned on review, long after the fact. They've still lost the game, and in reality, the penalty still stands.
harold · 15 June 2007
James Bond -
You seem to be missing a major point here.
Non-coding DNA is one of many massive problems for ID/creationism. Why would a designer use it every time he magically created a eukaryotic cell, if it was in any way imperfectly functional?
It is NOT a problem for science, nor is its lack - prokaryotes don't have it.
When creationists gloat over some suggestion of function in non-coding DNA, they're not scoring a point, they're merely trying to get one major (and valid) penalty against them overturned on review, long after the fact. It doesn't add a shred of support to their model if some of it has a function; they just think that it mildly reduces the impact of one of the thousands of pieces of evidence and logic against their model. They've still lost the game, and in reality, the penalty still stands.
someguy · 15 June 2007
Steve Reuland · 15 June 2007
CJO · 15 June 2007
It highlights a common thread in Creationist propaganda: The Big Lie.
Take the best evidence for evolution, play a game of "telephone" with it on the wingnut web, and then claim this garbled reiteration as a Creationist "prediction" to crow about with hollow triumphalism to the credulous and the already converted.
The Cambrian explosion? Selection-driven adaptive radiation into formerly unoccupied (well --non-existent, really) niches.
Irreducible Complexity? Basically evidence of a stochastic process at work, since the hallmarks of rational design at its best are simplicity and elegance, NOT needless complexity.
Complex Specified Information? Okay, Bill, explain this specification to me again. Sounds like function. Selectable function.
And now "junk" DNA. Utterly devastating, so the only way out is the Big Lie.
Steve Reuland · 15 June 2007
someguy · 15 June 2007
harold · 15 June 2007
Wade · 15 June 2007
Popper's ghost · 16 June 2007
Popper's ghost · 16 June 2007
Marc Geerlings · 16 June 2007
>I think very little unused sequences looks very suspiciously like ther fully
>functional genome predicted by ID.
>NOTHING is predicted by ID. Try to get that through your martini-addled brain, >Mr. Bond.
You're right nothing is predicted by ID, however I will predict that there are two scenario's exist that will give ID a major boost if they ever will become true:
1. DNA which is fully functional without a lot of "junk".
2. "Junk" which gets incorporated into the working part, which will be seen as the "creator" had put all our functionality all ready in the DNA and it gets used when needed.
Your comment, how ever true it is, does not help to see in advance how research is gonna used against science. Better to be prepared then willfully ignorant of the danger!
harold · 16 June 2007
harold · 16 June 2007
Marc Geerling -
You seem to be missing a major point here.
Non-coding DNA is one of many massive problems for ID/creationism. Why would a designer use it every time he magically created a eukaryotic cell, if it was in any way imperfectly functional?
But is NOT a problem for science, nor is its lack - prokaryotes don't have it. Science can explain non-coding DNA, ID can't, but take away the non-coding DNA and there are still millions of things that science can explain, and that ID not only can't, but doesn't even try to. Science doesn't rely on non-coding DNA, science was going strong before non-coding DNA was discovered. But non-coding DNA devastates ID.
When creationists gloat over some suggestion of function in non-coding DNA, they're not scoring a point, they're merely trying to get one major (and valid) penalty against them overturned on review, long after the fact.
It doesn't add a shred of support to ID if some or all non-coding DNA has some "function"; it just mildly reduces the impact of one of the thousands of pieces of evidence and logic against their model. Without non-coding DNA, ID makes no sense, with non-coding DNA, it makes no sense, but in the second case, there are .00001% more ways to explain why it makes no sense.
(See my post above for a partial explanation of why, even some of it is transcribed at a low level, most non-coding DNA probably doesn't have the same kind of strong, specific functionality as genes.)
Steve Reuland · 16 June 2007
Marc Geerlings · 16 June 2007
@Harold
I'm not missing to point, I agree completely with you that everything can explained by ID, but this is not a scientific battle. And Although ID doesn't predict anything scientifically, it is front-loaded.
- Everything is created by a perfect creator.
- The creation is perfect and by creation they are only thinking about humans.
So they make prediction, not based on science at all, but that is no problem for them, because the majority of the people think if you put on a lab-coat and say a lot of science jargon, that is science. And if something that was called junk which builds human, isn't junk after-all, this will be seen as a win for Jesus. I will give them wriggle room and maybe enough to convince a judge who is a little bit sympathetic. Again I'm not missing to point, I I only not convinced of the rationality of the majority of people in this world. I'm only saying try to think as the majority of people and especially as the people who try to get us back to the dark ages and have the power to do it. ID doesn't explain anything scientifically but I think majority of people think it does.
harold · 16 June 2007
Marc Geerlings -
I have a slightly optimistic reply to that.
One thing you may be experiencing is that a lot of people have a misunderstanding of the term "intelligent design".
A lot of intelligent people initially think that it means what it sounds like it means - they more or less mistake it for simply accepting science without ruling out some kind of God, along the lines of Ken Miller.
Once they find out that ID means such things as claiming that the bacterial flagellum could not have evolved, but had to be specifically, magically created by a "designer", or claiming that since we can tell that a beehive was designed by bees, therefore we should conclude that all living cells were "designed" by an "intelligent designer" that we ostensibly don't know anything about, they tend to reject it.
Only time will tell, but the battle against ID has been relatively successful so far.
I also wrote a post about why non-coding DNA isn't likely to be a major factor in body plan development. For some reason, it seems to have disappeared (it was extemely dry and scientific, not "abusive"). Hopefully it will reappear.
Torbjörn Larsson, OM · 16 June 2007
harold · 16 June 2007
Tobjorn Larsson -
I realize that hard-working volunteers run this site, and all I do is show up and comment once in a while, but still...
In fact, when I write anything inolved here, I usually copy it and paste into a text file, having been burned so many times (a much higher burn rate for attempted posts than any other site I am aware of).
Ironically, that "first time" message reassured me that the post had been registered, and I failed to do so.
It's interesting to wonder what "first time" means. I took it to mean "first time you have posted on this particular thread". Of course, that's a somewhat inefficient standard for initiating a review, since it will force a review of regular posters' contributions over and over again, for every new thread they post on.
An alternate explanation is that it's meant to identify true first time posters, but worked wrong and misidentified us as first-timers. A true first time post would be a logical standard for review. Of course, that only works if you correctly identify first time posts.
So it's either an inefficient screening system that worked correctly (it was our first post to this thread), or a potentially efficient screening system, but one that didn't work correctly.
For the record, I only support "banning" anyone either for repetitive disruptiveness long after their points have been addressed, or for really serious abuse, such as actual practical threats, intense profanity, or the use of offensive ethnic, gender, or orientation slurs, or something equally offensive.
Otherwise, I say, let the trolls be heard - and answered.
Steve Reuland · 16 June 2007
Harold, I published your comment, it should show up now. I have no idea why it was held as pending as I'm not the one who runs the site. I should have gotten an email about it, but either I didn't get it or I didn't notice it (which happens when you get 50+ emails per thread).
raven · 16 June 2007
Torbjörn Larsson, OM · 17 June 2007
raven · 24 June 2007