Home » 'Junk DNA', Darwinism » Are Selfish Genes Selfish? Are Retro-transposons Junk?

Are Selfish Genes Selfish? Are Retro-transposons Junk?

At PhysOrg.com, they have an article dealing with the CTCF protein and its binding sites. It turns out that the CTCF has both binding sites that are common to all mammalian lineages, and thus “conserved” and “ancient”, as well as binding sites found only in particular lineages. The binding sites found only in particular lineages are embedded inside “retro-transposons”, which “use a copy-paste mechanism to spread copies of themselves throughout the genome.”

However:

The retro-transposon’s copy-and-paste behaviour has long been considered totally self-serving. However, the study showed that when a retro-transposon containing a CTCF-binding sequence spreads around a mammal’s genome, it can deposit functional CTCF binding sites in novel locations, altering the activity of distant genes.

Further:

We looked at six mammalian species representing primates, marsupials, rodents and carnivores, and discovered a simple mechanism that they all use to remodel their DNA . . . ”

Another day; another bad day for Darwinism. It looks like “junk” isn’t “junk” after all. FYI, here’s the link.

  • Delicious
  • Facebook
  • Reddit
  • StumbleUpon
  • Twitter
  • RSS Feed

53 Responses to Are Selfish Genes Selfish? Are Retro-transposons Junk?

  1. A bad day? Why? Does evolutionary theory care about the extent of junk in genomes? Up until 1980, junk was undreamt of. I suspect that Ohno’s 1980 paper would have been reported then as “a bad day for Darwinism” – all this junk and no selective explanation! 30 years later we get a little function for some previously misclassified junk – and it’s a “bad day for Darwinism”!.

  2. It’s a sign of how desperate ID proponents are for a confirmed ‘prediction’. Unfortunately, ID does not predict that junk DNA has a function. That would be true only if we assumed that the designer would not have put nonfunctional elements into his design. As ID proponents are fond of saying, intelligent design is not optimal design.

    They can’t have it both ways.

  3. Guys it’s a bad day for darwinism because it is another layer of information it cannot account for.

    Not that it matters because that is just another promissory note…

  4. Guys it’s a bad day for darwinism because it is another layer of information it cannot account for.

    Crikey. A bad day for music theory as well then.

    (‘True’ junk, incidentally, would fall under the Neutral Theory. This is a central part of evolutionary theory but not, for historic reasons, “darwinism”. So it’s actually a good day for darwinism if it has a function!).

  5. No one can pretend that evolutionists didn’t throw junk DNA in everyones face for decades and even give the impression that they had predicted it.

    Up until 1980, junk was undreamt of.

    But now that it’s been explained, after the fact of course, the theory is tied to all the junk producing mechanisms posited to have been acting over eons. ‘Non coding’ DNA must accumulate? So when non-coding DNA turns out not to be non-functional it is not consistent. While ID would predict degradation, evolution surely predicts more ‘junk’ and so as it decreases – bad for evo.

  6. Then ToE is darwinism, Neutral Theory, both, none?

  7. No one can pretend that evolutionists didn’t throw junk DNA in everyones face for decades and even give the impression that they had predicted it.

    Citation, please. There is a lot of apparent junk in most eukaryotic genomes and hardly any in most prokaryotic; masses in the onion and barely any in the pufferfish. I am not aware of any retrospective claim of prediction for any of this.

    But now that it’s been explained, after the fact of course, the theory is tied to all the junk producing mechanisms posited to have been acting over eons. ‘Non coding’ DNA must accumulate? So when non-coding DNA turns out not to be non-functional it is not consistent. While ID would predict degradation, evolution surely predicts more ‘junk’ and so as it decreases – bad for evo.

    One of the more convoluted ways to get to the slogan ‘bad for evo’ I have encountered. What’s this new find shifted the percentage by? 0.01%? 0.02%? To what extent do you think evolution ‘predicts more junk’? Another 100 bases? an ounce? a ton? Still, when the amount of DNA does not change, as here (just relabelling), that is irrelevant anyway.

    Here’s the news – evolutionary theory is broadly concerned with changes in genotypes. Some genotypes have phenotypes, which influence genotype survival, and some don’t. Neutral or non-existent phenotypes (eg individual ‘true’ junk sequences) and non-neutral phenotypes both evolve. The total cost of all DNA is a phenotype of sorts, which I predict limits its indefinite expansion. Evo fine.

  8. Then ToE is darwinism, Neutral Theory, both, none?

    ToE is all mechanisms by which replicators change in population frequency and in lineages, and the consequences of this for populations and lineages. Neutral Theory is evolution in the absence of selection – the background process.

    Darwinism can mean several things:
    1) Common Descent
    2) Gradualism
    3) Natural Selection as the prime reason populations change.

    Hardly anyone outside of the odd popular science writer calls themselves a “darwinist” these days. He did write in 1859, after all, and things move on.

    1) Common Descent must be extended to include non-vertical means of gene transfer
    2) Gradualism dominates, but larger changes are more prevalent than he was aware.
    3) Natural Selection is now seen as a bias within the background process (population sampling) rather than the central mechanism. It remains important, but Drift (ie chance) is much more important than was once thought. One reason for this is the prevalance of neutral sequences – eg junk. A bad day for darwinism when Drift came along, I can tell yer. Ah yes, I remember it well …

  9. ToE is all mechanisms by which replicators change in population frequency and in lineages, and the consequences of this for populations and lineages. Neutral Theory is evolution in the absence of selection – the background process.

    Although to my mind it makes more sense to combine them in a single concept, in which “selection” is simply a bias to the stochastic sampling of alleles in each generation, in favour of what increased reproductive success in the parent generation.

  10. No citation needed, we’ve had to endure evos hopped up on ‘junk’ for too long. Changes in genotypes is not what evolution is broadly concerned about, an imaginary and impossible type of change is what evolution is ‘concerned’ with. Evolution must predict ‘a whole lot’ of junk, if it made real predictions, but will accommodate any amount. I don’t think anyone can speak with authority on the inner workings of life, we’re at the observe and wonder stage at best.

  11. Changes in genotypes is not what evolution is broadly concerned about

    I’m afraid you are mistaken. You are telling ‘evos’ what it is they study?

    Genotype is all that is inherited (you may shout “epigenetics!”, but epigenetic factors ultimately trace back to genetic ones). And to declare that evolution is only interested in limited ‘types of change’ is hogwash.

    Genotypes are sorted (by selection) according to some phenotypic effect that they have on the lives of organisms built from them. To that extent, yes, phenotype is important. They are also sorted (by drift) completely independently of their effects on phenotype. Either way, what gets squirted into the next generation are genes, functional and nonfunctional alike.

    Evolution must predict ‘a whole lot’ of junk, if it made real predictions, but will accommodate any amount.

    It really, really, depends on circumstances. Sorry. Evolutionary theory is about the regularities dictated by the central processes (replication in a finite world). The causes of any differential are necessarily circumstantial. The distributions of junk across various taxa are raw data, not predicted results – though the existence of junk was indeed predicted by an ‘evo’ (Ohno). I wasn’t aware he had been in your face about it.

    What ‘real predictions’ do you have in mind? Lottery numbers? The winner of the next Derby? What predictions do other sciences make?

    “If I hold a rock out and let go, I predict that it will fall”.
    “Suppose you are deep in space with no planet or star anywhere in the vicinty”.
    “OK smart-arse, if I hold a rock out and there is a massive body nearby it will fall”.
    “OK, exactly where on the surface will it end up?”
    “Ummmm…”.

    “Aha! Your so-called science fails to make a prediction. Nya-nya-nya-nya-nya”

    It is not the fault of evolutionary theory that the entities produced by the stochastic processes of evolution are divergent wrt this or that characteristic. “If being a giraffe is so great, how come everything on earth isn’t a giraffe?”.

    Junk was ‘predicted’ by an evo. No-one believed him, because ‘evos’ felt, like ID-ers, that such wastage would not be retained. There was a debate. 30 years later, and refined by widely available genome databases, it looks like the ‘selectionists’ were wrong (“a bad day for evo!”). The greater part (ie >50%) of most eukaryote genomes appears to be nonfunctional. Allowing for cryptic function (and being 5 times more generous than I am inclined to be) it is unlikely that we will see functional sequence in humans getting past 25%.

    That is not based upon evolutionary theory, it is based upon functional analysis. I could be wrong. And so bloody what if I am? What would that mean for evolutionary theory (as opposed to Ohno’s theory)? Another bad day for evo, of course. How do we cope?

  12. Although to my mind it makes more sense to combine them in a single concept, in which “selection” is simply a bias to the stochastic sampling of alleles in each generation, in favour of what increased reproductive success in the parent generation.

    Absolutely. It’s all about sampling. When selection is ‘off’, it’s all drift. Turning up the selective ‘heat’ diminishes, but does not eliminate, the role of drift.

    The compartments are historic – Selection, then Neutral Theory, then Near-neutral theory. The reality is a continuum, dependent on the intensity of selection.

  13. Changes in genotypes is not what evolution is broadly concerned about

    OK, as an explanatory framework for changes in phenotypes, that makes some sense. But at base, the underlying mechanism by which phenotypes change is by population and lineage changes in genotype. That may be mediated via phenotype; it may not.

  14. “in which “selection” is simply a bias to the stochastic sampling of alleles in each generation, in favour of what increased reproductive success in the parent generation.”

    Then, there is always “natural selection? never genetical drift?

  15. No, the other way round: there is always drift, and sometimes bias to that drift, aka “natural selection”.

    The metaphor goes like this:

    The imagine a bottle bobbing about on a choppy lake. If there are no currents (and no prevailing wind direction), it will take a “random walk” – it is likely to end up far from where it was first dropped, but there’s no way of knowing where, apart from knowing that the longer you’ve left it, the further from the start it is possible for it to end up.

    Now add a current. The thing will still do its “random walk” but there will be a bias to the probability of where it ends up – it’s more likely to end up downstream than in any other direction. The weaker the current, the less the probability that it will end up downstream in preference ot any other direction. The stronger the current, the greater the likelihood it will end up downstream.

    Or, my more potent metaphor:

    Imagine losing a toddler in IKEA. The longer he’s missing for, the further he could have got to, but there’s no way of knowing in which direction he’s likely to be (although he’ll never be far from where he was half a minute ago).

    However, if there is a hot dog stand in the store, there’s more chance his next move at any given time will be in the direction of the hot dogs. So your best bet (though it is still a remote one) is to head there.

    The maze of aisles in IKEA are drift. The hot dog stand is natural selection. The toddler is the population.

  16. Chas D and E Liddle:

    Neutral theory was proposed by Kimura because of the high degree of polymorphisms found in proteins. Because of genetic load, directional selection could not be taking place at so many loci all at once.

    Kimura promoted the idea of neutral drift coupled to positive selection way back in the 80′s.

    To this point, this is all just history. Yet, the problem remains that of a realistic time-frame. Yes, neutral drift happens, but it happens extremely slowly, not allowing realistic time-frames for significant genomic change. [The "infinite allele" model of Kimura is based on the fact that if an allele (whatever exactly this means) mutates, that the probability of it mutating again is so remote that it's as if this allele goes on as it is forever.] And positive selection still has the problem of genetic load, a problem exacerbated by the greater degree of polymorphism discovered through WGA.

    I honestly believe that it takes blind faith on the part of evolutionists to believe that known Darwinian processes can account for macro-evolution. This is the challenge for Darwinism; and it can’t rise to meet it.

  17. Elizabeth:

    Here’s another metaphor.

    Imagine a computer programmer who, having built his software and the computer to run it on, writes a program to carry out a specific function. Then imagine this programmer writing other such programs, and then stringing them together so that in the end, an incredibly complex level of functioning is made possible.

    I like my metaphor better than yours.

  18. Chas D:

    Darwinism can mean several things:
    1) Common Descent
    2) Gradualism
    3) Natural Selection as the prime reason populations change.

    And being an “evolutionist” can also mean several things:

    1) Life evolved
    2) Life evolved through random processes
    3) Life evolved through random processes and common descent.

    Hardly anyone outside of the odd popular science writer calls themselves a “darwinist” these days. He did write in 1859, after all, and things move on.

    However, it was Darwin who proposed his putative “theory of evolution”. And to call yourself an “evolutionist” fails to distinguish between those who, unlike YEC, believe that life forms developed over a time frame witnessed to by the fossil record. This is the same failure that marks the imprecision of equivocating that “evolution is a fact” and “I believe evolution occurred through Darwinian processes.”

    IOW, “evolutionist” is too vague. Darwinist is much more specific and accurate, as it identifies the person’s reliance on random variation and selection (environmental change). An over reliance on neutral drift renders NS an almost invisible force, and so I’m not sure what that makes someone. I guess I’ll call them “drifters”! ;)

  19. IOW, “evolutionist” is too vague. Darwinist is much more specific and accurate, as it identifies the person’s reliance on random variation and selection (environmental change).

    IOW, you completely misunderstand the point. I wasn’t proposing an alternative, merely pointing out the inaccuracy of the “darwinist” label.

    You can call me what the hell you like, to be honest, but simply in the interests of accuracy, calling all advocates of ‘unintelligent’ evolution darwinists because it “identifies the person’s reliance on random variation and selection” is precisely why you CANNOT call all such people darwinists! Inasmuch as you see it as a political movement or a religion, or need a handy label to lump together the heathen herd, knock yourself out.

    “Evolutionist”, to me, is like “geneticist” – one who studies evolution: mechanism, cause, consequence. And I guess I am neither, professionally.

    But no-one is a “Newtonist”, or a “Mendelist” or “Morganist” or “Fisherist”. One immediately finds oneself wondering which of Darwin’s many views (some accurate, some not) one is being associated with. It provides a useful means for the lazy to attack the subject, I suppose.

    Common descent preceded Darwin, and Natural Selection was independently elucidated both prior to and contemporary with his formulation. He was uncannily prescient on a huge amount of the modern theory, but …

    Even in your terms, it is difficult to determine why the discovery of a previously unknown functional role for some instances of a transpositional sequence should be a bad day for those who consider the mechanisms underlying evolution to be neither goal-directed nor demanding of intelligent tweaking.

  20. I honestly believe that it takes blind faith on the part of evolutionists to believe that known Darwinian processes can account for macro-evolution. This is the challenge for Darwinism; and it can’t rise to meet it.

    To respond, I need your definition of “macro-evolution”. If you have already given it, apologies, but could you repeat it?

  21. Transposons carry within its DNA sequence the coding for two of the enzymes it requires to “jump around”, meaning your position doesn’t have any explanation for them beyond “looky thar at them tahr thingies!”

  22. Transposons carry within its DNA sequence the coding for two of the enzymes it requires to “jump around”, meaning your position doesn’t have any explanation for them beyond “looky thar at them tahr thingies!”

    Not one of most incisive pieces of argumentation, Joe. Viable transposons code their own transposition, ’tis true. That itself is curious. If they were ‘for the benefit of the genome’, one would not expect to see them acting like viruses. An organism-controlled transposable sequence could be operated on solely by genes outside it – the sequence itself does not have to possess any genes at all. And better it doesn’t, because these things can rapidly get out of hand.

    Yet it does have genes – those genes benefit, first and foremost, the transposon. Having jumped, provided it does not break, a viable transposon can jump again. And again. And again. Each new viable copy can spawn new copies – an exponential process. Sometimes – it could not be otherwise – a transposon lands in a gene. This is much the same as a mutation. The sequence added is pretty ‘random’. Sometimes the result is detrimental, sometimes it is neutral. And sometimes it is beneficial.

    But MUCH more important to note is that the vast majority of transposon sequences in the genome are broken. They do not have viable copies of the enzymes required to jump around – which is a damn good job because these things can spread like wildfire when they get out of hand. It is, on the surface, a crap, scattergun way of generating novelty. “RM+NS” cannot do nuffing, but spraying transposons round like schrapnel can?

    Even if active transposons serve a ‘purpose’, there is no reason to suppose that inactive ones do by association. Their purpose cannot be to transpose, since they cannot transpose. One could argue that damage to transpositional ability is a designed ‘fail-safe’ to limit over-infestation. But nothing is sweeping up this debris. Dead transposons are, as far as can be told, junk, whatever their active or gene-inserted cousins may be up to. And there are millions of ‘em!

  23. Chas D:

    Viable transposons code their own transposition, ’tis true. That itself is curious. If they were ‘for the benefit of the genome’, one would not expect to see them acting like viruses.

    Nice strawman.

    But MUCH more important to note is that the vast majority of transposon sequences in the genome are broken.

    They have done their job- finished, no need to jump around.

    But anyway how does your position explain transposons? Or was I right…

  24. PaV

    I honestly believe that it takes blind faith on the part of evolutionists to believe that known Darwinian processes can account for macro-evolution. This is the challenge for Darwinism; and it can’t rise to meet it.

    Do you mean ‘complexity’, or evolution above the species level? The latter is, to my mind, pretty easy. I do not perceive it as an exercise in blind faith – but then, how would I know? :0)

    I guess the simplistic answer would be: serial elimination of ‘old’ alleles and introduction of new, in tandem with population division. We can characterise the process of allele introduction (mutation) and elimination or fixation within a particular recombining population. For populations of realistic sizes, I am not aware of any analytical result based on realistic mutation rates and mathematically tractable processes of fixation under various selective differentials that would prevent indefinite evolution of a lineage at allele level (ignoring, for this purpose, the complexity of the phenotypes).

    If there was such a barrier, then I might need a leap of faith to surmount it. But I am not aware of such a barrier from any analysis of mutation rates and substitution models.

    Neutral substitution takes a long time, and for a given allele in a diploid population is 4Ne – ie an ‘ideal’ population of a million individuals will on average see fixation in 4 million generations – glacially slow it might seem.

    But the overall neutral substitution rate is just the mutation rate. The mutation rate is effectively a per capita rate – a million individuals (pop A) produce a thousand times as many mutations as 1,000 (pop B). Each mutation takes a thousand times longer to fix in pop A, but there are a thousand times more of them. So in fact, both populations are fixing alleles at same rate: the mutation rate. So it is something of an illusion of intuition to see the fixation of one allele as a bar to evolution in a multi-allele system.

    So, this process (to me) generates an inexorable quality to the change in a population. This is agreed, to a variable degree, by the ID advocate to operate within and around species level. We can diversify bears by this means, but cannot diversify bears from raccoons, for example. But I don’t see where that comes from, barring intuition, which can be a shaky guide. Mutation fixation only occurs within a reproductive population. It derives from layered stochastic processes – mutation, drift and selective pressures are all stochastic. Once you have two isolated populations (isolation itself being a stochastic sampling process), you have populations that no longer keep each other in lock-step by recombination. They are genetically independent, and the only future I can see for them (without intervention or some unseen barrier to divergence) is ongoing, indefinite divergence. It is not clear that bear divergence is of an essentially different character from bear clade/raccoon clade divergence.

  25. Pav,

    (Chas) IOW, you completely misunderstand the point.

    Rereading, my response comes across as somewhat snarkier than intended or merited. So, I apologise!

  26. Re the concept of the Selfish Gene:

    I do recommend this lecture as an antidote:

    http://videolectures.net/eccs07_noble_psb/

    I keep recommending it, but haven’t got any feedback yet!

    I think you guys would like it.

    Don’t be put off by the fact that I do :)

  27. They have done their job- finished, no need to jump around.

    And therefore they are ex-entities. They are non-functional. They serve no purpose. They are, in a word, junk.

    But anyway how does your position explain transposons?

    Well, I could have sworn I addressed that. Transposons are selfish genetic entities whose transposition is a consequence of their ability to hijack the host’s mechanisms and transpose. By creating multiple copies, each with transpositional ability, such a sequence can spread. In that respect, they act, as I said, “like viruses”. And that resemblance is more than superficial. Read up on them; there is an extensive literature.

  28. Chas D:

    And therefore they are ex-entities. They are non-functional. They serve no purpose. They are, in a word, junk.

    That could be, that they are now junk.

    Transposons are selfish genetic entities whose transposition is a consequence of their ability to hijack the host’s mechanisms and transpose.

    That doesn’t tell us how they came to be.

  29. To say that genes are selfish, one needs 1 Dawkins of imagination. Inasmuch as evolution is postulated to have no intent or goal, it cannot be a cause of formal processes of any kind, and of recording prescriptive information in the form of DNA/RNA in particular. The presence of bona fide control is a very reliable (massively empirically warranted) indicator of choice contingency, not of necessity and/or chance.

  30. That could be, that they are now junk.

    Wow! A concession! I will chalk that up in my diary.

    That doesn’t tell us how they came to be.

    A HUGE amount of work has been done on how they came to be. There is a substantial amount of sequence overlap between the elements of various families of transposon and viruses. The traffic is probably two-way – there are transposons that create a capsid, as if to become an infective virus and leave the cell, but they do not, in fact, leave the cell, but transpose elsewhere within it. Such an element is a short evolutionary step away from becoming an infective virus. Other elements are more closely related to ‘true’ viruses than to anything in the host – they would appear to be viruses that have lost the ability to survive externally, but not to integrate and excise from genomes.

    Of course, you dismiss evidence based on common sequence as evidence of common design, so I’ll leave you to figure out how this all fits together on that hypothesis.

  31. To say that genes are selfish, one needs 1 Dawkins of imagination. Inasmuch as evolution is postulated to have no intent or goal, it cannot be a cause of formal processes of any kind, and of recording prescriptive information in the form of DNA/RNA in particular. The presence of bona fide control is a very reliable (massively empirically warranted) indicator of choice contingency, not of necessity and/or chance.

    I think you miss the point of the metaphorical use of the word “selfish”. An entity that has the capacity to spread will spread. If it does so at the expense of a larger unit – a host genome, say – it is reasonably termed “selfish”. We might also apply the term to a cancer, (which ultimately dies along with its host, so it’s selfish and short-sighted!) or a hookworm. Genetic parasites are parasites nonetheless.

  32. Thanks Elizabeth. Keeps stopping on my computer, but I’m working through.

    Interestingly, Burt and Trivers’ (rather good) Genes in Conflict uses the term “Selfish Genetic Element” (which I borrow) but they explain the distinction between theirs and Dawkins’s use of the term.

    They are specifically applying the term to genes that spread by subverting (another metaphor) the ‘conventional’ mechanisms of spread to gain an edge – usually by a distortion of the 50/50 Mendelian transmision ratio, but also by other familiar intragenomic mechanisms such as cut-and-paste ahead of the replication fork, and copy-and-paste in all its many forms. That is: genetic parasites, in sum. (Such parasites can become ‘domesticated’ of course).

    Dawkins was talking about an even more metaphorical ‘contest’ between alleles competing for a locus in a population via their effect on fitness. Still, Drive (distortion of 50/50 transmission) is a more direct version of that same competition – the only time rival alleles for a locus get a clear ‘shot’ at each other is during the diploid phase.

    I would hope that even people who struggle with Dawkins’ metaphor would recognise the appropriateness of the term ‘selfish’ for intragenomic competition, particularly where the elements decrease fitness.

  33. There’s also a book (The Music of Life) but the content of the book is pretty well in the lecture, so it makes sense ot link to that.

    It’s a nice book though.

    As he says, he is not contradicting Dawkins – he’s simply showing that you can turn the picture upside down, and it makes as much, if not more, sense.

  34. There’s also a book (The Music of Life)

    Ah, he wrote that? I have seen it on the shelves, and considered buying it but … didn’t. Something in my somewhat reductionist (albeit multi-level) mind baulked at it. I am a ‘genes-outward’ kind of guy, despite the phenotype/complexity bias that might come from actually being mostly diploid soma. I was also bugged that he’d nicked one of my (unpublished) metaphors. I’ll give him a fair go, though!

    As he says, he is not contradicting Dawkins – he’s simply showing that you can turn the picture upside down, and it makes as much, if not more, sense.

    He’s turning the picture its original way up again! I was very taken with a remark in a 1976 biochemistry lecture about this bloke who’d just written a book arguing that organisms were for DNA’s benefit, not the other way round. In the context of how things were seen up until then, it was a very interesting perspective shift. But of course one should not be dogmatic about perspective.

  35. Interesting!

    I think I would perceive the distinction as between someone (Dawkins) who is interested in the evolution of complex phenotypes and someone (Noble) who is interested in the operation of a complex phenotype. Both are interested in genes, but at a somewhat different level: germ line vs soma.

    Dawkins would see the succession of a germ line as the main stuff of interest from an evolutionary pov. It is that DNA that persists through the generations. And in multicellular organisms, that DNA is largely quiescent. Germ line DNA doesn’t build hearts, or make camouflage, or go through the many complexities of what Dawkins calls the ‘digression’ of a complex, multi-year somatic phase. It just sits there.

    But of course somas are built from the same DNA, and the interactions of the genes within such somas are grotesquely complex. In that sense – in ‘pulling out’ complex phenotype from genotype – reductionism hasn’t got a hope, but I don’t think that’s where Dawkins was focussing.

    Each life is a kind of empirical test of a particular genome. (In sexual species, it is the only time that genome is ever tested). It is only in the context of that empirical test that gene actions and interactions can be evaluated. But it almost does not matter that emergent actions are complicated, from an evolutionary perspective. One could construct an evolutionary process from genotypes alone, provided that such an artificial ‘life-form’ was somehow sustained. And I think that’s where Dawkins’s reductionism was aimed: at the unifying simplicity of underlying process – the evolutionary process.

    But of course ‘real’ life sustains itself (given basic inputs). Instead of existing solely as genotype, genes construct structures and processes that assist their persistence. The ability to contribute effectively to successive lives is the genetic quality that is being ‘assessed’ by the evolutionary process, in a very roundabout way. But I don’t think Dawkins was ever suggesting that complex somas are best understood from the bottom up. Somas are built from the same genes, but one process is only interested in the genes themselves, the other in their effects.

    I’m aware that genes are also expressed in germline cells, of course. But in a way, the germline cell is itself part of a ‘somatic’ construct, ‘for the benefit of’ germline DNA.

  36. Macroevolution is moving beyond the species level. For concreteness sake, I’m thinking of the Cambrian Explosion, where almost all present age body plans came about. Also, the transition from reptile to bird, and from reptile to mammal, etc. IOW, major phylogenetic change. This is way beyond known Darwinian mechanisms.

  37. Chas D:

    I guess the simplistic answer would be: serial elimination of ‘old’ alleles and introduction of new, in tandem with population division. We can characterise the process of allele introduction (mutation) and elimination or fixation within a particular recombining population. For populations of realistic sizes, I am not aware of any analytical result based on realistic mutation rates and mathematically tractable processes of fixation under various selective differentials that would prevent indefinite evolution of a lineage at allele level (ignoring, for this purpose, the complexity of the phenotypes).

    This is the leap of faith I’m talking about.

    If there was such a barrier, then I might need a leap of faith to surmount it. But I am not aware of such a barrier from any analysis of mutation rates and substitution models.

    According to Darwin, no such barrier exists. According to neo-Darwinists of today, no such barrier exists. But facts speak differently.

    I believe it was Michael Lynch or Allen Orr that did some theoretical study which showed that the first three mutations affecting some existing protein have impact, but that after the third mutation, nothing much happens. This bespeaks a limit. Huxley was troubled by the fact that species boundaries could not be hurdled. You take a cat, inbreed it all you want, and you get a fair amount of change; but then the inbred cats become sickly. It bespeaks a limit (barrier). Same thing with dogs.

    Look at Behe’s Edge of Evolution, where the malarial parasite, in a battle for its very life, can only come up with about two or three amino acid changes (consistent with Lynch/Orr).

    To “believe” that you can move beyond species barriers in any fundamental way (I’m not talking about adaptive radiation) takes just that: Belief.

    The kind of plasticity that Darwin was convinced of, simply doesn’t show its face.

    But the overall neutral substitution rate is just the mutation rate. The mutation rate is effectively a per capita rate – a million individuals (pop A) produce a thousand times as many mutations as 1,000 (pop B). Each mutation takes a thousand times longer to fix in pop A, but there are a thousand times more of them. So in fact, both populations are fixing alleles at same rate: the mutation rate. So it is something of an illusion of intuition to see the fixation of one allele as a bar to evolution in a multi-allele system.

    In my view, this is ignorance parading itself as insight.

    What is missing from you analysis is the reality that as one genome sweeps its way through the population, it’s wiping out huge amounts of neutral drift present in the other genomes within the population. You can’t have your cake, and eat it too. To have any one genome be the possessor of more than perhaps ‘two’ beneficial mutations sweeping themselves to fixation becomes hugely improbable. So fixation has to take place pretty much in ‘step-wise’ fashion. And this brings in the 4Ne problem.

    Now Elizabeth doesn’t accept this. She prefers to ‘believe’ that ‘alleles’ (whatever that means) become ‘fixed’ as needed. I can’t seem to convince her that for more than one or two changes sweeping to fixation at once, any neutral mutations present within the population is available to the population as a whole can only be helpful to the population if it first becomes fixed. Kimura did a calculation showing that at a particular locus, 3 million mutations took place before a particular mutation became fixed at that location. Mutations are one thing; beneficial mutations are another; and having that mutation come to the aid of the population is still another.

  38. Chas D:

    IOW, you completely misunderstand the point. I wasn’t proposing an alternative, merely pointing out the inaccuracy of the “darwinist” label.

    You seem to have misunderstood my point.

    If we look at the fossil record, it is quite clear that “evolution” has taken place. We see reptiles arising on land with stunning diversity. The same for birds and mammals. Obviously there has been some ‘progression’ of forms. To deny this is to fall in the YEC camp.

    I don’t.

    So, I agree that “evolution” is a fact. So how does calling yourself an “evolutionist” distinguish you in any way from my position?

    Now, if you believe that RM+NS leads to the progression of forms we see, then you and I disagree. You want to call me an IDer, fine. I’ll call you a Darwinist. He was, after all, the one who proposed that which you believe in.

    Even in your terms, it is difficult to determine why the discovery of a previously unknown functional role for some instances of a transpositional sequence should be a bad day for those who consider the mechanisms underlying evolution to be neither goal-directed nor demanding of intelligent tweaking.

    Here’s why:
    (1) It more strongly confirms that putative “junk-DNA”, used by Darwinists as a (theological) argument against ‘design’, is actually NOT junk. Darwinism strikes out.
    (2) It is obvious that some kind of mechanism—unknown to, and not complementary, to neo-Darwinian mechanisms is at play. This only undermines the credibility of neo-Darwinism. As I’ve pointed out at this blog before, I’ve stopped studying population genetics since it is obvious that it is all so passe.

    What are Darwinists left with? Maybe James Shapiro can come to the rescue. But in embracing him, you’re going to have to let go of Darwin almost completely. Where does that leave you?? Here: We know what happens, but we don’t know how it began happening. Having put Darwin to the one side, then I think we should go back to where science was before Darwin veered us all of onto the wrong track. Right?

  39. Maybe James Shapiro can come to the rescue.

    I can’t imagine how or why, since nothing he says is in disagreement with mainstream biology.

    Actually he seems to take the same position that Elizabeth takes, which is that evolution is a kind of intelligence.

    He could be right or wrong about the emphasis he places on certain kinds of processes involved in variation, but he is entirely mainstream in saying that variation does not track or foresee need, that it is “random” with respect to need.

  40. Petrushka:

    . . . but he is entirely mainstream in saying that variation does not track or foresee need, that it is “random” with respect to need.

    I’m not very familiar yet with Shapiro’s work, but I don’t think this captures his thought well. If you talk about “engineering”, then there has to be an element of teleology in it, wouldn’t you agree?

  41. Chas D:

    Thanks for the expression of civility. Our world needs it!

  42. No.

    Engineering is a metaphor. Shapiro is quite clear that variation is not foresightful.

  43. The phrase he uses is, “future adaptive needs are unknowable.”

    He does claim that the mechanism for generating variation is similar to the mechanism that generates antibodies, that is to say it generates more sophisticated mutations than point mutations.

    But it’s a difference in emphasis rather than a claim that mainstream biology is wrong. He doesn’t list any mechanisms not known in the mainstream.

  44. Now Elizabeth doesn’t accept this. She prefers to ‘believe’ that ‘alleles’ (whatever that means)

    Alleles are variant forms of genes. If a gene has “polymorphism” i.e. exists in several variants, each of those variants is called an “allele”.

    become ‘fixed’ as needed. I can’t seem to convince her that for more than one or two changes sweeping to fixation at once, any neutral mutations present within the population is available to the population as a whole can only be helpful to the population if it first becomes fixed.

    No. An allele, by definition, is not “fixed”. If one allele became fixed (i.e. possessed by every member of a population) there would be no variants. You seem to have fundamentally misunderstood my point.

    Kimura did a calculation showing that at a particular locus, 3 million mutations took place before a particular mutation became fixed at that location. Mutations are one thing; beneficial mutations are another; and having that mutation come to the aid of the population is still another.

    And this is absolutely not what I was suggesting (nor what any evolutionary biologist is suggesting).

    I’ll try to explain later, but this is, simply, wrong.

    Cheers

    Lizzie

  45. Here’s why:
    (1) It more strongly confirms that putative “junk-DNA”, used by Darwinists as a (theological) argument against ‘design’, is actually NOT junk. Darwinism strikes out.

    Darwinism (sigh!) is actually neutral on the role of active designers (the science, not the people doing it). It just does not require them to mediate change in populations. Those Darwinists who use junk as an argument against design might find that argument (marginally) weakened, but ultimately, Darwin theorised in blissful ignorance of genetics, let alone DNA. There are many ‘naturalistic evolutionists’ who think that there is a lot more function in DNA than, say, Larry Moran. (“Than Larry Moran thinks”, not “than there is in Larry Moran”!). So in the absence of ID theory, it would be a bad day for one set of Darwinists vs another; the presence of ID in the mix does not cause function to devalue the whole naturalistic theory just because one of the objections to ID is countered. A good day for selectionist Darwinists. And in fact, a good day for anyone who is interested in new discovery.

    The thing is, I think, that people are not arguing for junk on the basis of dogma, or because it makes ID look silly. They might agree with Ohno’s argument, but must dispense with it if data demonstrates that it is invalid. But half the sequence in (say) the human genome consists of inactive transposons. That is simply a piece of data, not a prediction of Darwinism. To call that junk is not an ideological position. Relatives of these sequences may exist in functional code, but to declare these non-junk by association is like saying there is no unemployment because your brother just got a job. No-one is arguing there is no functional sequence, and no-one is arguing that it is all functional sequence. There is a notional ‘slider’ somewhere between 0% and 100% junk. I place it, on the basis of what we know about the junk sequence and its features, at around the 90% mark. If function is in fact discovered in the ‘debris’ classes, that’s good – scientists are interested in discovery. Would my worldview crumble? Hardly! And who would make such a discovery? My bet would be: a Darwinist.

    (2) It is obvious that some kind of mechanism—unknown to, and not complementary, to neo-Darwinian mechanisms is at play. This only undermines the credibility of neo-Darwinism.

    neo-Darwinian mechanisms of mutation, recombination, drift and selection operate upon junk just as much as non-junk. The assumption that junk should be eliminated because it is so costly is an intuition shared by ID and many in the ‘selectionist’ camp, but it betrays a failure to grasp the relevance of population genetics. It will only be eliminated if individuals that have eliminated it are significantly better off (fitter) than those that haven’t. Junk can accumulate by a ratchet, or by transposon infestation, in tandem with a bias against random deletions (the only mechanism to get rid of it). If the selective disadvantage of each particular extension is minimal (and with 3,000,000,000 base pairs in the population, most typical extensions would be near-invisible), the intuition based on overall cost is misplaced.

    What are Darwinists left with?

    The same as what they started with!

  46. OK, in brief:

    Let’s take a few regulatory genes that have an effect on, say beak size. Let’s say three genes. Each of these have several alleles. Some alleles tend to promote larger beak size, some, smaller. All these alleles are found in a finch population.

    If a finch happens to inherit alleles that make for smaller beaks then it will have a smaller beak than average. Ditto larger. But most finches will inherit a cocktail of alleles of the three genes that gives them an average beak size. In other words the central limit theorem ensures that beak size will have a Gaussian distribution, as there are more ways of getting a mid-size beak than a small or a large one (like tossing three dice).

    Now, change the climate so that there are a reduced number of small seeds and an increased number of large ones.

    The alleles that tend to promote smaller beaks will become rarer, and the alleles that tend to promote larger beaks will become more common. Average beak size in the population increases.

    Meanwhile, new alleles are constantly being created for these three genes. Some will result in smaller beaks, some larger.

    Over time, if the climate change remains, some alleles that promote small beaks may actually drop out of the population, and be lost forever. Meanwhile, any new alleles that tend to promote larger beaks will tend to propagate quite rapidly and become quite prevalent.

    That’s what I mean by a constant drip-feed of new alleles into the population, giving rise to variance that is then culled by the current environment, resulting in adaptive change.

    I hope that is clearer.

    “Fixation” is a tricky concept, and better, IMO, considered in its antithesis – the idea that some alleles can vanish from a population.

    Fixation of alleles is rarely total, but extinction of an allele is.

  47. What is missing from you analysis is the reality that as one genome sweeps its way through the population, it’s wiping out huge amounts of neutral drift present in the other genomes within the population. You can’t have your cake, and eat it too. To have any one genome be the possessor of more than perhaps ‘two’ beneficial mutations sweeping themselves to fixation becomes hugely improbable. So fixation has to take place pretty much in ‘step-wise’ fashion. And this brings in the 4Ne problem.

    What was that about ignorance and insight?

    A genome only sweeps its way to fixation in an asexual population. In a sexual population, recombination occurring anywhere in the DNA strand swaps genes. The result of this slicing and dicing is that each locus can be considered independently. Sex has huge consequences for populations and their ability to explore phase space. Every one of a population of a million has the potential to be the ‘discoverer’ of a new mutation at a single locus. Every locus is subject to this same ‘search’. To get two mutations into the same individual, you do not have to fix them in the population first. The population is not the unit of interest for ‘benefit’. The FIRST individual with a beneficial mutation benefits from it – before anyone else in the population has it. If it becomes fixed, that individual is the ancestor of everyone at that locus. Other loci come from other ancestors. A sexual population can be in the process of fixing multiple alleles simultaneously, at different loci. It can even create new genes with two mutations from two singly mutated ones, by tacking the back end of one to the front of another.

  48. “Fixation” is a tricky concept, and better, IMO, considered in its antithesis – the idea that some alleles can vanish from a population.

    I think it helps to consider it as fixation or extinction of ancestry – coalescence, effectively, looking backwards. It may well be that a particular sequence rarely remains unchanged from start to finish of the fixation process, but there is a point at which every member of the population can trace its family tree for that locus back to a single individual, and all other ancestors are ‘extinct’ at that locus.

    Fixation of alleles is rarely total, but extinction of an allele is.

    Well … surely fixation must have been total for all non-polymorphic alleles? Which account for most of a species’s ‘canonic’ genome?

  49. 49

    PaV,

    It is obvious that some kind of mechanism—unknown to, and not complementary, to neo-Darwinian mechanisms is at play.

    Will you be publishing this remarkable insight, perhaps in one of the ID supporting journals.

    If you can actually support this claim with evidence then it would cause a firestorm in the “Darwinist” world, regardless of what journal it was published in.

  50. Trufax :)

    But then a non-polymorphic gene wouldn’t have alleles :D

    Point taken though.

    cheers

    Lizzie

  51. But then a non-polymorphic gene wouldn’t have alleles :D

    Nnnnngggghhh! Damn this slippery language! Of course, ‘wild-type’ regains allele status every time a mutation occurs, however sickly and doomed its possessor! :0)

  52. :D

    Actually there is a real problem here that we have to be constantly aware of: words like “allele” and “gene” are just approximations – models of reality that we use to categorise nature, and nature is never as tidy as our models.

    I think there is a danger of over-reifying these concepts and forgetting that they just reflect ways of thinking about thing, and that different ways of thinking may shed different insights.

  53. I think there is a danger of over-reifying these concepts and forgetting that they just reflect ways of thinking about thing(s)

    Yes, if we could strip away all the organismal stuff and just track DNA molecules in space, we’d see a whirring cloud of these evanescent linkages of bases, colliding, recombining, replicating, dissipating. It would be an unholy mess.

    But fortunately they oblige by generating more persistent and sufficiently discrete and classifiable units that we can keep track of – organisms, gametes, genes, populations etc. Even so, trying to follow all that activity, conceptually, down the generations, can be like trying to follow a bee in a swarm, or a prawn from plankton to sandwich.

    One problem is that all of these mappings are a manifestation of phenotype, and the most basic processes of “DNA management” – DNA replication, repair, mitosis, syngamy, meiosis and recombination – have no access to broader phenotype. They slice straight through our ‘units’ as if they weren’t there. Because to them, they aren’t. And it is at the ‘raw’, unprocessed DNA level that evolution (change) actually occurs.

Leave a Reply