Philip Ball is a freelance science writer based in London (UK). He frequently writes for Nature. His latest article is a review of a recently published paper by John Avise [What a shoddy piece of work is man]. Apparently Avise has just published a paper in PNAS where he points out that our genome does not look like it was designed. It's an attack on Intelligent Design Creationism and Adaptationism. I can't find the paper but I have read Avise's book, Inside the Human Genome so I'm familiar with his thesis—and I agree with it.
The purpose of this posting is not to review the points that John Avise makes but to comment on one of the points made by Philip Ball. At the end of his Nature review he says,
However — although heaven forbid that this should seem to let ID off the hook — it is worth pointing out that some of the genomic inefficiencies Avise lists are still imperfectly understood. We should be cautious about writing them off as 'flaws', lest we make the same mistake evident in the labelling as 'junk DNA' genomic material that seems increasingly to play a biological role. There seems little prospect that the genome will ever emerge as a paragon of good engineering, but we shouldn't too quickly derogate that which we do not yet understand.THEME
Genomes & Junk DNA
I just gave a talk on junk DNA where I explained to my audience the nature of the scientific controversy. We know for a fact that our genome is littered with pseudogenes of all sorts and we know for a fact that more than 50% of our genome is repetitive DNA of one kind or another. A good hunk of that is degenerative transposons and fragements of transposons [Junk in your Genome: LINEs]. Another large hunk is Alu sequences: fragments of an ancient primate transposon derived from 7SL RNA [Transcription of the 7SL Gene].
We also know a great deal about introns and that knowledge leads to the conclusion that most intron sequences are dispensable. it's part of the junk in our genome. We know about the genetic load argument [Genetic Load, Neutral Theory, and Junk DNA] and we know about the C-Value Paradox. Most scientists who study the problem of junk DNA know about The Onion Test.
My point is that it's extremely misleading to suggest that our identification of junk DNA is based on a lack of understanding. That's simply not true. There are some very good scientific reasons for maintaining that most of our DNA is junk based on over 40 years of work on genome organization. Yes, it's true that there have been some scientific challenges questioning the conclusion of those studies. There is a group of scientists who claim that vast amounts of our genome serve some mysterious purpose that's only vaguely defined. It could be regulation of some sort or even an entire new class of RNA-encoding genes that make us human.
These claims make the debate over junk DNA a scientific controversy but they certainly haven't succeeded in disproving the hypothesis. None of the recent claimants can explain pseudogenes and degenerative transposons, which make up more than half of our genome. None of the opponents can refute the genetic load argument.
Science writers like Philip Ball can be forgiven for not delving into the problem. It's easy to fall for the latest articles that purport to show function for a large part of what we call junk DNA. After all, those anti-junk proponents don't do their homework either and they gloss over all the data that contradicts their "new" hypothesis.
My point is that the idea of junk DNA is alive and well in spite of what modern science writers seem to think. It's just not true that today's scientists think we made a big mistake in the past by calling it junk DNA. This is still very much a scientific controversy and it's too soon to tell how it will pan out.
Personally, I think the evidence in favor of a large amount of junk in our genome is persuasive and I'd be very, very surprised if a significant amount of it turns out to be functional. I wish science writers would stop behaving as though the issue had been resolved and junk DNA is dead.
Agreed. Given how good we scientists are at dealing with quantitative data in the details of our and others' research, we're remarkably bad at seeing the big picture in its correct proportions.
ReplyDeleteIs there anyway to prove or show some (if not most) part of our genome is junk. Can we delete the junk part to observe for phenotype changes? Or have such experiments being conducted?
ReplyDeleteYou can't delete junk in humans for the simple reason that we don't do such experiments with humans.
ReplyDeleteThis comment has been removed by the author.
ReplyDeleteWho cares, human dna isn't a magical exception. Just use plants or flies or something.
ReplyDeleteWe already know there's variation in genome size within the same species. As i understand it there are differences on the order of several percent in the middle repetitive DNA of the drosophila genome and other measurable variations (cf DNA loss and evolution of genome size in Drosophila, Petrov (2004)).
Such variations could conceiveably have an effect on fertility against originals if the size differences prevent things from lining up, i suppose (even while retaining the same functions and forms).
There's already some interesting notes on plants btw (Rapid DNA loss as a counterbalance to genome expansion through retrotransposon proliferation in plants, Hawkinsa (2009)): "Studies in maize suggest a doubling of its genome over as little as 3 million years due to TE accumulation alone (10, 11)."
It'd be great to be able to find a working ancient sample that can be fertilized with a modern strain.
There's been at least some work in doing large scale deletions in mice. Here's one abstract I could find:
ReplyDeleteNature. 2004 Oct 21;431(7011):988-93.
Megabase deletions of gene deserts result in viable mice.
Nóbrega MA, Zhu Y, Plajzer-Frick I, Afzal V, Rubin EM.
DOE Joint Genome Institute Walnut Creek, California 94598, USA.
Abstract
The functional importance of the roughly 98% of mammalian genomes not corresponding to protein coding sequences remains largely undetermined. Here we show that some large-scale deletions of the non-coding DNA referred to as gene deserts can be well tolerated by an organism. We deleted two large non-coding intervals, 1,511 kilobases and 845 kilobases in length, from the mouse genome. Viable mice homozygous for the deletions were generated and were indistinguishable from wild-type littermates with regard to morphology, reproductive fitness, growth, longevity and a variety of parameters assaying general homeostasis. Further detailed analysis of the expression of multiple genes bracketing the deletions revealed only minor expression differences in homozygous deletion and wild-type mice. Together, the two deleted segments harbour 1,243 non-coding sequences conserved between humans and rodents (more than 100 base pairs, 70% identity). Some of the deleted sequences might encode for functions unidentified in our screen; nonetheless, these studies further support the existence of potentially 'disposable DNA' in the genomes of mammals.
Of course, even these megabase deletions represent only ~ 0.1% of the total mouse genome, so I'm not sure you can really extrapolate very far.
The negative result of the mouse deletion experiments aren't all that meaningful. There are plenty of examples of deleting well-established functional genes from mice that also don't cause any noticeable phenotypes. A positive result from the gene desert deletion, on the other hand, would have been interesting.
ReplyDeleteI can totally imagine some IDers/Creationists spinning the genetic load argument in their favor. God put all that junk DNA in there as a buffer to protect us from random deleterious mutations. See, it's design!
ReplyDeleteFirst comment to that Nature piece:
ReplyDelete"In your DNA you have your program, your recipe, your code made up of a code of four bases adenine, cytosine, guanine and thymine. Instead of the 2 value O's and 1's of Binary, a 4 value code of A,C,G,T"
Oh. My. God. That hit right at the heart of one of my worst pet peeves ever. The "DNA=program" delusion. Fucking. HAAAAAAATE. That analogy. SO MUCH.
Ok, I'm done frothing at the mouth now, but seriously: there is just so much wrong with claiming the genome to be a base-4 version of computer software. The genome is nothing like software. There is a whole world of parallel epigenetic, including extragenomic, inheritance out there. Lipids, for one, are vertically inherited, and thus can be argued to bear 'information' (eg. a la Beisson & Sonneborn 1965 PNAS), thereby rendering the whole idea of 'program', 'code' and much of simplistic information theory completely useless for any real biology.
On that note, the genome isn't a 'blueprint' either. See, the harder you look, the less and less any of it looks 'designed', and the more you wonder if there could be anything stupider and less efficient than the process of evolution...were it to have any goal to begin with.
But I seriously wish people would stop with the 'software delusion'. Grrrrr....
I don't particularly like the term "junk DNA" because it connotes that such DNA is uninteresting. I used to think junk DNA wasn't interesting at all but have become more fascinated with it with time. I don't know what else I would call it but I don't like the name "junk."
ReplyDeleteNon-coding DNA?
ReplyDeleteHow about "vestigial DNA"?
ReplyDeletePausanias,
ReplyDeleteIf you were referring to my comment, I don't think non-coding DNA is sufficient. Doesn't non-coding DNA refer to both junk DNA and regulatory sequences? And perhaps I'm wrong, but aren't retrotransposons normally considered junk?
anonymous,
ReplyDeleteI can totally imagine some IDers/Creationists spinning the genetic load argument in their favor. God put all that junk DNA in there as a buffer to protect us from random deleterious mutations. See, it's design!
Actually there are "serious" scientists who have proposed just that! (Without the God.)
I always ask them to explain how having extra junk DNA protects against DNA replication errors in the genes but they never seem to have an answer.
It's a really stupid argument whether it's made by IDiots or people pretending to be scientists.
anonymous says,
ReplyDeleteThe negative result of the mouse deletion experiments aren't all that meaningful. There are plenty of examples of deleting well-established functional genes from mice that also don't cause any noticeable phenotypes. A positive result from the gene desert deletion, on the other hand, would have been interesting.
The experiment was an attempt to falsify a hypothesis; namely, the hypothesis that the deleted DNA is junk.
It failed in its primary objective. This does not prove that all junk DNA is really junk but it does lend some credibility to that idea, especially since the regions of the genome were deliberately chosen because they include regions that produce long non-coding RNAs.
We now have some evidence that these RNAs are not required.
It's the nature of real scientific controversies that no single experiment is going to persuade anyone tho change their minds. There will always be ways to rationalize any result that goes against what you believe.
I have no trouble doing this for any so-called "evidence" purporting to show a function for most of our genomic DNA. My opponents are equally versatile. :-)
Dunbar says,
ReplyDeleteYou can't delete junk in humans for the simple reason that we don't do such experiments with humans.
But God has no such restrictions. She repeatedly plays around with our genome by chopping out bits here and there and inserting other bits at random. That's why the human population contains so many copy number variants and deletion/insertion variants.
The parts that are missing in some of us are good candidates for "junk," right?
Unless, of course, you believe that those of us who are missing parts of our genome are also missing something else. :-)
Riffing on the "function of junk": Could the repetitive, random, or non-functional stuff--instead of protecting protein-coding and regulatory DNA from mutation--rather increase variability of offspring?
ReplyDeleteIIRC, the likelihood that two neighboring genes (and of course other relevant sequences) get passed on to the F1 is dependent on crossing-over, which is dependent on the distance between the genes. More junk in between thus means greater likelihood of new gene combinations. Could this be a way to modulate the recombination of traits?
In other words, does the onion sometimes need greater variability in trait recombination, perhaps as response to its ecological situation? (I guess this would make serious sense only for eucaryotes but could well have arisen as a function of the genome with early procaryotes.)
[Sorry for the incoherence / possible ignorance. Genetics has been a while ago for me.]
@Psi Wavefunction
ReplyDeleteWhile I'd agree that our analogies don't always bear up in every last detail (they're analogies FFS, they don't have to!), I think that the DNA = program one is pretty damned apt. The genetic code is a close relative of EBCDIC and ASCII in symbolic representation terms; control sequences are logic gates (not just 'like' logic gates - they integrate signals to give a True/False answer); processes may display a classic modular sequence/choice/iteration structure. I would be more tentative about using the term 'information', but I recognise the temptation.
Extragenomic inheritance is just something else, just as a television with a chip in it is something else. But as far as the fundamentals of the genome are concerned, I would favour computing analogies. And even more annoying ones - literature, or music. If I ever wrote a book, you'd be for lobbing it across the room! But it's all about providing appropriate pegs for hanging concepts upon. Dry, analogy-free science makes for dull journalism.
Larry Moran,
ReplyDelete"anonymous,
I can totally imagine some IDers/Creationists spinning the genetic load argument in their favor. God put all that junk DNA in there as a buffer to protect us from random deleterious mutations. See, it's design!
Actually there are "serious" scientists who have proposed just that! (Without the God.)"
I want to ask these people why bacteria didn't evolve a buffer and instead have such compact genomes.
BB suggests,
ReplyDeleteIIRC, the likelihood that two neighboring genes (and of course other relevant sequences) get passed on to the F1 is dependent on crossing-over, which is dependent on the distance between the genes. More junk in between thus means greater likelihood of new gene combinations. Could this be a way to modulate the recombination of traits?
What you're suggesting is that there was selection for extra DNA between genes in some species in order to increase the variation in future generations. How did that work? The mutations that expanded the genome happened gradually in individuals - how did those mutations "know" that they were going to be useful in the future even though they had no effect on the individual?
That's not how natural selection works.
Besides, the best way to enhance gene shuffling is to have more chromosomes. If what you're suggesting is true then there should have been an increase in the number of chromosome over time. The last thing you'd expect would be a decrease in the number of chromosomes but that's exactly what happened in the human lineage.
@Allan Guess I despise the program analogy so much because I'm a proud cell biologist and not a molecular geneticist ;-)
ReplyDeleteTo me, it is very important to emphasise that there is more to oranisms than their genetic code. Much more. And in my completely unbiased opinion here, the truly interesting stuff happens outside the genome =P
Also, the logic gate analogies are a bit bleh because unlike electronic systems, biochemical interactions are so many orders of magnitude messier it doesn't even really make sense to apply those comparisons even at the most basic level. Anyone who's worked with gene interactions in vivo would probably sigh heavily about now. I'm all for analogies, but I think the DNA=blueprint/program thing does more harm than good by this point. In a way, the public has been a little too heavily 'programmed' with that one!
How about 'recipe for proteins'? Or 'a program for making proteins'?
@Psi
ReplyDeleteGuess you won't be interested in my forthcoming book "Life is Just One Big Computer Program So There", then? :0P
@Allan Guess I despise the program analogy so much because I'm a proud cell biologist and not a molecular geneticist ;-)
Ah well, y'see, I'm a (ex) molecular biologist and a computer programmer ... but I'm interested in 'nature' also.
It's a question of the level you're referencing. We are tapping away to each other at our keyboards without any need to consider that underlying every keystroke (especially [SEND]) is a mess of digital wizardry. Ditto biology. But if you're talking about details in the way DNA works, I reckon computing is as good as any - with the proviso that your readers may be no wiser about how computers work than they are about the genome.
I'm not a fan of the "recipe" angle, though!
I'm partial to a musical one - DNA the sheet music dots, but subject to the mercies of the player, the acoustics, noises off ... yeah, I know: DNA dynamics resemble nothing more closely than DNA dynamics! :0)
In Light of Evolution IV: The Human Conditions Sackler Colloquium: Footprints of nonsentient design inside the human genome.
ReplyDeleteAvise JC.
Proc Natl Acad Sci U S A. 2010 May 5. [Epub ahead of print]
Larry: can you recommend any particular review articles about the controversy over junk DNA? I'm inclined to agree with the pro-junk side, but I don't know nearly enough about the details of the evidence, so an overview would be very helpful if such articles exist.
ReplyDelete