Quantcast

«

»

Dec 07 2013

The reification of the gene

Razib Khan poked me on twitter yesterday on the topic of David Dobbs’ controversial article, which I’ve already discussed (I liked it). I’m in the minority here; Jerry Coyne has two rebuttals, and Richard Dawkins himself has replied. There has also been a lot of pushback in the comments here. I think they all miss the mark, and represent an attempt to shoehorn everything into an established, successful research program, without acknowledging any of the inadequacies of genetic reductionism.

Before I continue, let’s get one thing clear: I am saying that understanding genes is fundamental, important, and productive, but it is not sufficient to explain evolution, development, or cell biology.

But what the hell do we mean by a “gene”? Sure, it’s a transcribed sequence in the genome that produces a functional product; it’s activity is dependent to a significant degree on the sequence of nucleotides within it, and we can identify similar genes in multiple lineages, and analyze variations both as a measure of evolutionary history and often, adaptive function. This is great stuff that keeps science careers humming just figuring it out at that level. Again, I’m not dissing that level of analysis, nor do I think it is trivial.

However, I look at it as a cell and developmental biologist, and there’s so much more. That gene’s transcriptional state is going to depend on the histones that enfold it and the enzymes that may have modified it; it’s going to depend on its genetic neighborhood and other genes around it; it’s not just sitting there, doing its own thing solo. And you will cry out, but those are just products of other genes, histone genes and methylation enzymes and DNA binding proteins, and their sequences of nucleotides! And I will agree, but there’s nothing “just” about it. Expression of each of those genes is dependent on their histones and methylation state. And further, those properties are contingent on the history and environment of the cell — you can’t describe the state of the first gene by reciting the sequences of all of those other genes.

Furthermore, the state of that gene is dependent on activators and repressors, enhancer and silencer sequences. And once again, I will be told that those are just genetic sequences and we can compile all those patterns, no problem. And I will say again, the sequence is not sufficient: you also need to know the history of all the interlinked bits and pieces. What activators and repressors are present is simply not derivable from the genes alone.

And I can go further and point out that once the gene is transcribed, the RNA may be spliced (sometimes alternatively) and edited, processed thoroughly, and be subject to yet more opportunities for control. I will be told again that those processes are ultimately a product of genes, and I will say in vain…but you don’t account for all the cellular and environmental events with sequence information!

And then that RNA is exported to the cytoplasm, where it encounters other micro RNAs and finds itself in a rich and complex environment, competing with other gene products for translation, while also being turned over by enzymes that are breaking it down.

Yes, it is in an environment full of gene products. You know my objection by now.

And then it is translated into protein at some rate regulated by other factors in the cell (yeah, gene products in many cases), and it is chaperoned and transported and methylated and acetylated and glycosylated and ubiquitinated and phosphorylated, and assembled into protein complexes with all these other gene products, and its behavior will depend on signals and the phosphorylation etc. state of other proteins, and I will freely and happily stipulate that you can trace many of those events back to other genes, and that they respond in interesting ways to changes in the sequences of those genes.

But I will also rudely tell you that we don’t understand the process yet. Knowing the genes is not enough.

It’s as if we’re looking at a single point on a hologram and describing it in detail, and making guesses about its contribution to the whole, but failing to signify the importance of the diffraction patterns at every point in the image to our perception of the whole. And further, we wave off any criticism that demands a more holistic perspective by saying that those other points? They’re just like the point I’m studying. Once I understand this one, we’ll know what’s going on with the others.

That’s the peril of a historically successful, productive research program. We get locked in to a model; there is the appeal of being able to use solid, established protocols to gather lots of publishable data, and to keep on doing it over and over. It’s real information, and useful, but it also propagates the illusion of comprehension. We are not motivated to step away from the busy, churning machine of data gathering and rethink our theories.

We forget that our theories are purely human constructs designed to help us simplify and make sense of a complex universe, and most seriously we fail to see how our theories shape our interpretation of the data…and they shape what data we look for! That’s my objection to the model of evolution in The Selfish Gene: it sure is useful, too useful, and there are looming barriers to our understanding of biology that are going to require another Dawkins to disseminate.

Let me try to explain with a metaphor — always a dangerous thing, but especially dangerous because I’m going to use a computer metaphor, and those things always grip people’s brains a little bit too hard.

In the early days of home computing, we had these boxes where the input to memory was direct: you’d manually step through the addresses, and then there was a set of switches on the front that you’d use to toggle the bits at that location on and off. When a program was running, you’d see the lights blinking on and off as the processor stepped through each instruction. Later, we had other tools: I recall tinkering with antique 8-bit computers by opening them up and clipping voltmeters or an oscilloscope to pins on the memory board and watching bits changing during execution. Then as the tools got better, we had monitors/debuggers we could run that would step-trace and display the contents of memory locations. Or you could pick any memory location and instantly change the value stored there.

That’s where we’re at in biology right now, staring at the blinking lights of the genome. We can look at a location in the genome — a gene — and we can compare how the data stored there changes over developmental or evolutionary time. There’s no mistaking that it is real and interesting information, but it tells us about as much about how the whole organism works and changes as having a readout that displays the number stored at x03A574DC on our iPhone will tell us how iOS works. Maybe it’s useful; maybe there’s a number stored there that tells you something about the time, or the version, or if you set it to zero it causes the phone to reboot, but let’s not pretend that we know much about what the machine is actually doing. We’re looking at it from the wrong perspective to figure that out.

You could, after all, describe the operation of a computer by cataloging the state of all of its memory bits in each clock cycle. You might see patterns. You might infer the presence of interesting and significant bits, and you could even experimentally tweak them and see what happens. Is that the best way to understand how it works? I’d say you’re missing a whole ‘nother conceptual level that would do a better job of explaining it.

Only we lack that theory that would help us understand that level right now. It’s fine to keep step-tracing the genome right now, and maybe that will provide the insight some bright mind will need to come up with a higher order explanation, but let’s not elide the fact that we don’t have it yet. Maybe we should step back and look for it.

90 comments

Skip to comment form

  1. 1
    Simon Hayward

    Do I hear the distant thunder of theist hordes….charging forwards in an attempt to move this argument up another couple of levels.

  2. 2
    PZ Myers

    Won’t help. Any higher level explanation still needs to rest on a material, natural, genetic foundation.

  3. 3
    Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden

    I read Coyne, and his Lamarckian interpretation of Dobbs was well received by his commenters. I still think it was off base – when I read it, it didn’t seem Lamarckian at all (parents sprint more b/c they have to even though they aren’t in shape for it, kids sprint more b/c following parents and so are in shape for it before they get to adulthood and are better at it, grandkids sprint more b/c they are following parents who were better at it than the grandparents were, thus same dynamic, but with better teachers). But if you have a whole comment pool not getting it, then it’s clear that there are a large number of people out there with no training at all that won’t get it.

    Their efforts won’t work with a large value of “us” (Coyne has good reason to reject Lamarckism and isn’t going to flounder when theists drop by spouting nonsense, inspired by Dobbs or not, Dobbs won’t be taken in, and neither PZ nor his Horde will be taken in) but the “us” is ultimately finite: there are too many out there who don’t know anything about evolution and Gish and would-be-protege WLC and similar types make clear that they’ll grab anything, understood or not, in their wild attempts to fling different shit at the wall until something sticks.

    So, yeah, Simon Hayward is right.

    PZ is right that it won’t change the science on the ground [and I'm NOT saying scientists should stop having debates b/c they might be mischaracterized by the ridiculous], but I think it’s a very reasonable prediction to say that this will be picked up by the Gishes and WLC’s of the world.

  4. 4
    PZ Myers

    Yeah, the “lamarckian” nonsense is just a put-down, and anyone who throws the charge of lamarckism at these concepts doesn’t understand them.

    I’m also a little annoyed at the frequent claim that it’s “just” epigenetics, and it doesn’t matter because it’s only going to count if you get long-term heritable changes. Epigenetics matters! It just doesn’t matter in the sense a lot of people want it to — it’s another case of trying to shoehorn a significant modulator of gene expression into a model that only recognizes genes.

  5. 5
    eigenperson

    I think PZ is saying that one cannot understand an organism by understanding its genes. Coyne is arguing that in theory, one could understand an organism by understanding everything about its genes.

    I don’t think these are contradictory statements.

    The reason they are looking at the same article so differently is that Coyne saw it as a completely unwarranted attack on the idea that genes are the almost-exclusive-but-not-quite method of information transfer between generations. That’s how I read the article too, so I was inclined to agree with Coyne.

    Of course PZ is right too, but this particular article seems to make a much more sweeping (and less justified) argument than PZ does. At least, that’s what I got out of it.

    As a non-expert, I don’t know the status of West-Eberhard’s theory, or if either Dobbs or Coyne represents it accurately.

  6. 6
    ChasCPeterson

    This is all well and good as a developmental/physiological perspective, it’s useful and interesting and good, but if you think it reflects the emphasis and message of Dobbs’s piece then you read a different piece than I did. You seem to be projecting your own useful and interesting and good perspective on something that (I agree with Coyne and Dawkins here) was a sensationalized muddled mess.

    To address your point (rather than Dobbs’s, whatever that might have been), it’s the old push-n-pull of reductionism and holism.
    [for teh lurkers: Biolological phenomena occur and can/must be usefully and interestingly and well studied at many different hierarchically-arranged levels of organization--molecule, organelle, cell, tissue, organ, system, organism, population, community to name some. Phenomena at any one level find their causal mechanistic explanations at lower levels (reductionism, working dwon the hierarchy by breaking wholes into constituent parts, the strategy that has produced pretty much all of our knowledge of physiology, development, molecular bio, etc.) but their significance and contextual function at higher levels (holism, working up the hierarchy to discover emergent properties that depend on the organization of the parts in space and time).]
    I see your OP as calling for a more holistic view of genes functioning in complex integrated organizations in space and time. And of course I would agree that such a view is necessary for a deeper understanding of development and physiology and the evolution thereof. But as you point out, we do not have this kind of deep, integrative understanding of these extremely complex phenomena, and it will be a long time before do, if we ever do. There is therefore no warrant (yet, if ever) for abandoning the reductionist approach, and trumpeting “die, selfish gene, die!” is just wildly premature..

    Look: if it’s biological, it evolved. If it evolved, it’s heritable. If it’s heritable, it’s genetic. And if it’s genetic then yes, I must insist that it does in fact reduce to sequence information. The fact that that information functions at higher integrated levels in crazy-complex ways does not invalidate the reductionist view. And our present understanding of biological evolution (not real-time physiological or developmental or ecological function) is still most usefully conceptualized and studied at the level of genes. We can get to the emergent shit later. Maybe. I hope.

  7. 7
    PZ Myers

    If you put it this way,

    I think PZ is saying that one cannot understand an organism by understanding its genes. Coyne is arguing that in theory, one could understand an organism by understanding everything about its genes.
    I don’t think these are contradictory statements.

    They are contradictory. And further, that version of Coyne is wrong.

    We wouldn’t argue that if we understood every atom in the human body, we’d understand poetry, after all.

  8. 8
    Kausik Datta

    My understanding of what PZ is saying is that, simply put, the whole is greater than the sum of its parts, and therefore, the understanding of the whole cannot be easily approached using a reductionist approach. Reductionism and pattern-seeking, of course, have their respective importance in the process of trying to cobble together an understanding of the system as a whole – but emphasizing solely upon reductionist approaches render those efforts incomplete. Am I getting this horribly wrong?

  9. 9
    ChasCPeterson

    the “lamarckian” nonsense is just a put-down, and anyone who throws the charge of lamarckism at these concepts doesn’t understand them.

    The charge wasn’t thrown at the concepts, it was thrown at Dobbs’s specific stupid heuristic example, which (charitably) presupposed sufficient plasticity that two generations could produce an entirely different-looking animal from training alone, without acknowledging that plasticity itself has to evolve and therefore has genetic bases.

  10. 10
    New England Bob

    This post is the final straw for me. PZ, you have been going off the deep end for months. Your posts are no longer informative or even pleasurable to read. I am now taking your blog out of my RSS reader. I know you don’t care and neither do I any longer.

  11. 11
    PZ Myers

    New England Bob, #10:

    It must have been all the feminism in the post. Or maybe the atheism. Or maybe it was just the meanness.

  12. 12
    eigenperson

    They are contradictory. And further, that version of Coyne is wrong.

    I don’t think they are contradictory. The “in theory” is important. Perhaps I’m misrepresenting both you and Coyne, though.

    No, you probably can’t understand poetry from looking at just the human genome — or, better, you certainly can’t predict the Complete Works of Shakespeare from reading Shakespeare’s genome. To do that you would need to know everything about Shakespeare’s environment down to the atomic level, which is impossible.

    But I don’t think anyone is trying to use biology to predict someone’s Complete Works from their genome. The argument is more about how to explain whether a caterpillar is green or black. There is a gene-environment interaction here, but it’s not something that requires an atomic-level knowledge of the environment to understand.

    I think everyone, including you, agrees that heat-induced greening in caterpillars is mediated by genes, and that the evolutionary changes in those caterpillars occurred through selection on those genes. The difference seems to be that Coyne is fine with saying “Look, you agree that genes are at work here. So there’s nothing special going on — maybe it’s more complicated than we’re used to, but eventually we’ll understand the network of genes well enough to explain all the possible gene-environment interactions.” On the other hand, you don’t believe that we will ever understand the network of genes well enough to do that (because the system is far too complex and the very idea of considering “all the possible gene-environment interactions” is ludicrous).

    But, as I said above, maybe I don’t understand what either you or Coyne is saying.

    It must have been all the feminism in the post. Or maybe the atheism. Or maybe it was just the meanness.

    No, it was the science. You see, we don’t want you to talk about science. That doesn’t make us feel special and superior. Can’t we go back to the days when you stabbed crackers and we could laugh at silly people instead of having to think? [/sarcasm]

  13. 13
    Stephen Casper

    I really appreciate this discussion. I’d like draw your attention to Hans-Jörg Rheinberger’s book “An Epistemology of the Concrete: Twentieth-Century Histories of Life (Experimental Futures: Technological Lives, Scientific Arts, Anthropological Voices)” and particularly where he discusses the long history of the concept of the gene (Chapter 8). You may know it, and apologies if you do. Rheinberger discusses in this chapter the conceptual changes genes underwent in the twentieth century- from classical to molecular genetics and from molecular genetics to genomics by the century’s end. I think many of the elements of your essay here – and at least for me confusing elements of this history of science – derive from the rather difficult conceptual translation of genes through that history of molecular biology into the contemporaneous and parallel history of evolutionary science. Putting it too crassly: I find it hard to think about the gene as anything but a set of epistemological claims in rapid flux of which the “selfish gene” happens to be a brilliant but nevertheless limited subset (and a subset that also happens to be now an object of history as well).

  14. 14
    Rob Grigjanis

    To an ignorant outsider, this sounds like ‘Chemistry isn’t Physics’ in another language, and without the Chemistry. In other words, there is a need for higher-level principles?

  15. 15
    davidwhitlock

    Excellent explanation.

    Let me try to explain it a different way.

    Books are comprised of pages comprised of sentences composed of words comprised of letters.

    This is analogous to how organisms are comprised of cells comprised of gene products composed of genes comprised of nucleotides.

    The theme of the book is an emergent property of all of the pages and is not derivable from the words.

    The phenotype of an organism is an emergent property of all of the cells and is not derivable from the genes.

    The difference between Moby Dick and Gone With The Wind is the arrangement of the words. All of the letters (nucleotides) are the same. Virtually all of the words (genes) in both works are the same. The themes of the books (phenotype) are very different.

    You can’t study the themes of the two books by focusing on the words.

    An example of a “selfish gene” is a Homing Endonuclease Gene. An analogy would be a “selfish phrase”, that managed to get copied into a book and could never be erased. Sort of like a “typo” when books were replicated by being hand-copied. A phrase that hijacks the copying mechanism to preferentially replicate itself; a “make an extra n copies of this phrase” phrase. Books that acquired this “selfish phrase” would then be adversely affected depending on how many copies were transcribed in the next iteration of the book. If n=10, then in 5 “generations” there are 10^5 copies of the phrase. If the total book size is limited to 1,000 pages of 1,000 words, then 10^5 copies of “make an extra 10 copies of this phrase” would constitute 80% of the book (8 words times 10^5 copies). The “theme” of the book would become uninteresting and no one would bother copying it again. The book would go “extinct” because of the “selfish phrase”.

    This is why we don’t see any selfish genes of the type that Dawkins predicted. Real organisms with selfish genes that can take over the whole genome go extinct. A Green Beard gene is susceptible to parasitism by a Green Beard mimic, a gene that expresses a Green Beard but which does not incur any Green Beard associated costs (the cost of killing non-Green Beards).

    I suspect that this is what limits the fidelity of eukaryotic DNA replication. The fidelity can’t be arbitrarily high, or homing endonuclease genes would take it over. There has to be enough error in DNA replication so that can’t happen.

    Dawkins hypothesized that if a gene resulted in an association of light skin and behaviors favoring others with light skin, that such a gene could become dominant in a population. This is not the case because such a gene would be out competed by a gene that simply provided light skin without any of the favoring behaviors (which impose a cost).

    PZ is exactly right. A gene-centric approach is too simplistic and it is blinding researchers to other stuff that is going on.

  16. 16
    wolfgang

    I apologise in advance, since I am trying to point out something very simple no one (so far) has formulated very clearly (as far as I can tell).

    A. Dawkins and Coyne state that the behaviour of cells is a (very nice, complicated, complex, interwoven, far from completely understood) consequence of some very elementary rules on the level of elementary units (genes, or molecules and atoms – depending on how far one wants to go) and their environment.

    B. PZ points out that such a reduction to elementary rules and units is (probably) theoretically correct, but often not *useful* to derive the properties of a given cell in a given environment. In that practical sense, the reduction is not a satisfying explanation.

    Unless I am misrepresenting the authors, I do not see why these points should be incompatible at all. I am almost too afraid to compare it with other areas in science, for the same reason PZ pointed out in the post above, but here I go anyway …

    A’. The weather and climate are, as far as we understand, a (mathematically interesting, chaotic, difficult) consequence of some elementary rules (of physics and chemistry) acting on some elementary units (molecules). I do not think many people would disagree with that …

    B’. … but from a practical aspect, this approach is quite limited in its predictive power. That’s fine! We have other models, that are not as fundamental in the same sense, and which are remarkably useful.

    There is no conflict. Another example:

    A”. Much of the world we interact with every day is a (wonderful, … etc) consequence of the rules of quantum mechanics acting on elementary particles.

    B.” But from a practical point of view, these rules are as good as useless when trying to describe the behaviour of two colliding balls. That’s fine! We have Newtonian mechanics, which is not as fundamental as quantum mechanics, but fantastically useful for such practical applications.

    Again, there is no conflict. I apologise once more if I have represented the arguments of PZ (or Coyne/Dawkins/Pinker/…) incorrectly.

  17. 17
    ChasCPeterson

    This is why we don’t see any selfish genes of the type that Dawkins predicted. Real organisms with selfish genes that can take over the whole genome go extinct.

    the fuck? Have you read past the title?

    Dawkins hypothesized that if a gene resulted in an association of light skin and behaviors favoring others with light skin, that such a gene could become dominant in a population. This is not the case because such a gene would be out competed by a gene that simply provided light skin without any of the favoring behaviors (which impose a cost).

    You’re assuming an intrinsic benefit to light skin alone as well as the presence of a particular competing allele.
    I think. Actually it’s pretty tough to figure out what you’re talking about.

  18. 18
    franko

    My twopennyworth. When cells divide, or male and female gametes fuse, they carry into the progeny not just the DNA but all the associated non-DNA bits and pieces that happen to be around at the time. How the progeny cell(s) develop(s) is therefore a consequence not just of the replicated DNA sequence but also of all the other stuff. Therefore the unit that behaves “selfishly” (as in Dawkins’s model) is not solely the DNA. Even in the simplest case we know of — viruses — the DNA (or RNA) is invariably accompanied by proteins and other materials.

    But so far as we know, the least changeable of all the biological components is the DNA. Extensive mechanisms exist to maintain copy fidelity of DNA sequences. This suggests the DNA sequence is probably the most important part of the whole caboodle, but the entity that emerges is bound to be modulated by the other components (calling them epigenetic is a good handle for the time being). If this view of things is correct, then Dawkins, Coyne and Myers are also all correct. They’re all saying the same thing but with different emphases. The one who doesn’t look good is Dobbs, whose article has too much unscholarly journalese to be worthy of praise. On the other hand, he does seem to have provoked a lot of fascinating discussion…

  19. 19
    johnharshman

    PZ, you seem to be thinking about development while Jerry is thinking about evolution. All these epigenetic effects are important in development, but they aren’t what changes in evolution; that’s all genomic sequences. Dobbs is in fact talking about evolution and goes so far as to claim that the major differences between humans and cows are just epigenetic, while the sequence differences are both minor and unimportant. Can you possibly agree with that?

  20. 20
    Scientismist

    We wouldn’t argue that if we understood every atom in the human body, we’d understand poetry, after all.

    I have to disagree just a bit with PZ here, but only to support a more broad understanding of “understanding.”

    If we understood every atom in the human body, and how it got where it is, then we would indeed have to understand poetry (and a lot more); but we would not have understood much about the poetry by pointing to the dance of the atoms in the likes of Robert Browning’s brain; we might understand a bit more about it by studying something both more remote and more intimate, like the poetry of Elizabeth Barrett Browning.

    “Understanding” is always incomplete. In the example of cell biology (or bit-level computer science, in PZ’s analogy), understanding increases a bit when we can make sense of the effects between (among) levels. I had a student lab assistant once for whom I wrote a recommendation for a grad school program in material sciences. He later wrote to thank me, saying that in the first couple of weeks of the program he had learned more about how a transistor works than he ever wanted to know. A knowledge of how electrons dance through semi-conductive junctions might say a lot about why PZ got different voltages at different places and times with his volt meter; but it wouldn’t tell much about why the patterns dance in cohorts of eight at a time, since that is largely an historical artifact driven by the 256 values of ASCII. Complete understanding would have to cover the history of the English alphabet and the Arabic numbering system.

    In the undergrad cell biology course I took in about 1966, the professor delighted in pointing out to us that we would go through a different definition of “gene” for almost each week of the course (after almost 50 years, there are even more). That same prof arranged a bus to take us from our small liberal arts college to the big state university to see a seminar by Paul Boyer (who had been considered in the running for the Nobel prize, which had just been awarded to Jacob, Monod, and Lwoff for their work on the lac operon). Boyer at the time was working on ATP synthetase, which undergoes shifts in its subunits that are important to the sequence of catalytic events. In answer to a question, he referred to that “new French disease, called ‘Allosterie’” (allosteric interaction — another area of research for Monod — is a more general phenomenon in which an activator or inactivator ligand can shift the 3-D subunit structure of an enzyme, and affect its catalytic activity by binding to a site that can be far removed from the catalytic site). The lesson for us undergrads was not just that gene expression can be affected by mechanisms at many different levels (including both the internal and external environment of a large multi-subunit enzyme system), but that Nobel-level scientists can be just as snarky as the rest of us.

  21. 21
    grahamjones

    I suppose analogies are as risky as metaphors, but here goes. Like johnharshman, I’m more interested in the evolutionary consequences.

    Suppose you are observing a game in which the players each have some tokens which they move
    around on a board. You would like to understand the progress of the game, and ultimately to be
    able to predict how various aspects of the game will unfold. You are able to make some
    observations about the players’ moves and their consequences, and also able to listen to some of
    the players’ thoughts about their position in the game and their plans. Which is most important
    for the future of the game: the players’ moves or the players’ thoughts?

    Two extreme points of view would be:

    1. Only the positions of the tokens on the board count. The players cannot influence the board
    with their minds. Only when a player moves a token does anything `real’ happen. The players’
    thoughts may be interesting in their own right, but they are ephemeral and leave no permanant
    trace on the game.

    2. Every move in the game starts with a thought. The players may have poor assessments of their
    positions, or infeasible plans, but nothing happens without thought of some sort. The players’
    thoughts determine the progress of the game. The board merely records the players’ decisions.

    The analogy with evolution works like this:

    player – population
    token – individual
    position of a player’s tokens – gene frequencies in a population
    move – change of gene frequencies
    thought – developmental plasticity
    psychokinesis – Lamarckian evolution

  22. 22
    PZ Myers

    I’m talking about development and evolution. There certainly is far more to evolution than genomic sequences — those sequences need an environment (genes, cytoplasm, signals, etc) in which to operate.

    I took the cow statements as rhetorical flourish. But still, most of a cow is the same as a human and vice versa — it’s more than epigenetics, though, it’s changes in gene regulation that are inherited (and ok, some trans-acting changes).

  23. 23
    Pan Paniscus

    I’ve read Coyne’s two posts, Dawkins’s post, and now PZ’s above. It seems to me that all of them are right and all are consistent with each other on the science (perhaps with differing emphases at times).

    anyone who throws the charge of lamarckism at these concepts doesn’t understand them.

    That depends on what “these concepts” refers to. There is nothing Lamarckian in PZ’s post above. Whether there is in Dobbs’s article is unclear (most of the discussion on Coyne’s blog results from no-one being all that clear what Dobbs was arguing for).

    maybe that will provide the insight some bright mind will need to come up with a higher order explanation, but let’s not elide the fact that we don’t have it yet.

    I’m guessing that the “higher levels” are just layers and layers of complications, rather than the sort of thing that results in a neat explanatory theory.

  24. 24
    sawells

    I think this is actually the same problem we had with the futurology guys, who were arguing that the information content of the genome set some kind of limit on the complexity of the brain – as if the bitwise information content of a recipe was a measure of the complexity of the cake…

    Genes do nothing except in the context of the cell, and the genome does not contain enough information to reconstruct a cell – it doesn’t have to. Cells have been inheriting cell membranes, for example, in a continuous chain since before there were cells, so there’s no instructions in any genome for building a working cell membrane from scratch.

  25. 25
    patrick85

    Brilliant commentary all-around! What I have to add may not be what you had in mind, and, frankly, might not even be that well received – but I think it’s worth trying to get across (no matter how convoluted and clumsily I do it). And while it may appear broad, it rings true right down to particulars.

    Firstly, to echo your point, an examination of particulars is important and the current dialogue is certainly worth having. As Aldous Huxley notes:

    “The world is a continuum; but in order to act upon it successfully, we have to analyse it into easily comprehensible elements. The cake of experience can be cut in many different ways, and none of the systems of slicing can express the molar fact completely; each, however, may be useful for some particular purpose. There have been literally hundreds of analyses of human nature, some excellent, others less good, others again positively misleading.”

    So, genes, the genome, plenotypes, epigenetics, the microbiome, the “external” environment…all factors of the continuum worthy of analysis. But, I think it’s important to fundamentally recognize and remember – and I may well be accused of “too damn” high-order or even wishy-washy thinking – that, to paraphrase Alan Watts, the world doesn’t come “bitted,” and we won’t fully grasp its processes with a linear mode of thought. There has been note of this and some have pointed to Dawkins’ “The Extended Phenotype” and how genes “respond to” and/or “influence” their environment, but it extends beyond that in this way:

    Self implies other. Or, to put it in relevant terms, selfish implies (and involves and includes) everything else there is. In other words, it’s foolish to think in terms of “selfish” without a re-examination of “self” – which ultimately shows us that the world is really full of self-others as a continuum. In the same way, every gene “in an environment” is in all actuality a gene-environment. It doesn’t just “respond to” or “influence” it, it ultimately and fundamentally is it. Every single “one.”

    For example, one might readily acknowledge the interdependent relationship between bees and flowers. They might also be willing to take it a step further and recognize that interdependence truly suggests they are one continuous organism – for where there are no flowers there are no bees, and where there are no bees (or a similar pollinating organism) there are no flowers. They go-together in the same way a head goes with feet on a human body. So the “bee genome” really involves and includes the “flower genome.”

    In somewhat the same way, we’re all very familiar with the idea of having “internal” organs, but hardly anyone recognizes on our dependence on “external” organs. Half of our lungs could be said to exist outside of us in the form carbon dioxide-inhaling and oxygen-exhaling trees (and similar plant life). So the “human genome” goes-with the “tree genome.” To paraphrase Watts again, rocks are just as much you as your fingernails. You need rocks. What are you going to stand on?

    But, then, it goes even further than that simple relationship and includes any and all other organisms. From pollinators, to seed dispersers, to fertilizing ruminants, to the microbes, insects, and nutrient recycling mycelium in the soil, to the sea and the ecosystem and hydrologic cycle that sustain it – it’s a mutually arising, supportive, dependent, and functioning process. Any seemingly “individual” organism is an aspect of one “continuous organism,” as it were:

    Organism-Environment, the transactional nature – Alan Watts (clip)
    http://youtu.be/nLI54vXxfic

    In other words, you don’t just inherent “your” genes when you are born, you inherent your entire environment (and all the genes therein that exist or have existed). We cannot separate isolated unities in these universal networks of interrelationships and interconnections. There are no “parts” really separated in any level of the evolution scale. On the contrary, as in a holographic plate each “fragment” of the world is not any other thing than a concrete expression of the same and unique totality (or, in short, an expression of totality).

    It all goes together – Watts (clip)
    http://youtu.be/qml1-xzPpxY

    How do we define ourselves? – Watts (clip)
    http://youtu.be/lXRPjdXGjjg

    The genes that exist today are not something that is cut-off and the result at the end of the process, they are still part the process that required any and all other genes to have existed and done their thing. There is no “selfish” without other, even those that seemingly “lost” the “competition” (especially those that lost, actually).

    So, while science has and continues to wonderful strides in the examination of evolution and rightly does so breaking it down to its various components – genes, the microbiome, and the like – I think it’s just as important to emphasize and facilitate a shift in the “sense of self/identity,” which might help us to work towards building better models. With one more parting quote from Watts:

    “Look, here is a tree in the garden and every summer is produces apples, and we call it an apple tree because the tree ‘apples.’ That’s what it does. Alright, now here is a solar system inside a galaxy, and one of the peculiarities of this solar system is that at least on the planet earth, the thing peoples! In just the same way that an apple tree apples, the earth peoples!”

    Alan Watts – The Myopic View Of The World (lecture)

  26. 26
    Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden

    Okay, I need someone with a PZ-like or maybe Chas-like (not sure where his expertise lies, precisely, though it’s far more biological than mine) understanding of genetics and transcription/expression.

    I’m thinking about these concepts and it reawakened a question I had for my 9th grade bio teacher that couldn’t be answered by him:

    What, precisely, makes a dominant gene dominant, and a recessive gene recessive?

    And, even more importantly, is there a mechanism that can make a D/R pair switch their hierarchical places?

    When talking about sweeping change in a population that happens in a brief period of time, if a less common and recessive gene provided some benefit, but its recessive nature was making penetration take a long time, and then a change happened – whatever, however, I have no idea – that caused this beneficial gene to function as dominant, it seems that would vastly encourage the spread of the gene.

    Am I wrong?

    Am I right, but this is impossible?

    And what’s going on here with the D/R thing anyway?

  27. 27
    Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden

    I has an html sad.

    Obviously I was supposed to close the blockquote. :cry:

  28. 28
    PZ Myers

    Most frequent cause: recessive alleles are loss-of-function mutations. These are broken genes. Any wild type allele will do their job, so wt is dominant.

    Dominant alleles may be gain-of-function mutations. They do something constitutively, or excessively. Paired with a wild type allele, they still carry on with whatever their new/unregulated function is, so you still see them.

    There are other ways, though, too.

    Also, dominant/recessive are always relative to a specific allele.

  29. 29
    Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden

    Also, dominant/recessive are always relative to a specific allele.

    I knew that part.

    So does that mean that in most cases there is no dominant/recessive relationship?

    Is it all about phenotype, then, and not about transcription?

    Hypothetical, since I’m really, really ignorant here & want to understand (remember, I’m trying to understand how D/R works, not how eye-color works):

    1. There’s a gene that produces a deep purple eye color, almost like smoke over water while a distant recording studio is on fire.

    2. There’s a gene that produces a bright red eye color, like the flames coming off a burning wall that used a specific metallic pigment in paint.

    3. When you have 2 bright reds, you get bright red eye color.

    4. When you have 2 deep purples, you get deep purple eye color.

    5. When you have 1 of each, both pigments are created by the cellular machinery, but the dark purple overshadows the red, and eyes look purple – they appear phenotypically indistinct from having 2 copies of the DP gene. The BR gene is actually expressed, but has no impact on perceived phenotype.

    6. We humans label DP dominant and BR recessive, but the cellular machinery doesn’t treat them differently in any way, it’s just that the effects of one are more observable.
    ================

    Is this a better understanding than I had previously, if previously I thought that there was a difference in how the machinery expresses the gene?

    Or is it that the “loss-of-function mutations…broken genes” statement is meant to lead me to understand something different? Like the transcription is happening fine, but the gene itself codes for a protein that frequently breaks down before it can have its effect?
    =================
    Final attempt at understanding this:

    gain-of-function could imply that what’s happening here is that a gene for eye color is actually a regulatory gene that could cause either red or blue pigments, or both, to be produced.

    1. DP does in fact cause both pigments to be produced, which, when combined in the iris, produce a deep purple color.

    2. BR causes only the red pigment to be produced, without the partnering blue pigment, red eyes result.

    3. When you have one copy of DP and one copy of BR, you still get both pigments produced, therefore you still get deep purple eyes.

    4. There is no dominance, but absent DP you get only BR, because there is as of yet no “navy blue” gene. If NB existed, you could get the phenotype DP by having one NB and one BR.

    5. Either NB or BR could be seen as loss-of-function in comparison to DP, but DP could also be seen gain-of-function compared to either NB or BR. Which was true is simply a matter of which gene arose first, a question of the particular history of the gene, but the ways in which a loss-of-function pair operates would look identical to the way in which a gain-in-function pair operates, and both might be described by something like my above.

    ================
    Again, way above my biological pay grade, so I’m conceding none of my offerings might hit the mark. I’m just trying to understand PZ’s #28.
    ================
    Finally, it seems that if I understand PZ, epigenetic or environmental impacts on expression may help make animals differ, but that wouldn’t occur by way of flipping a dominant/recessive relationship b/c D/R doesn’t work at all like how I thought it did.

  30. 30
    chris61

    Crip Dyke

    It is also worth noting that what makes an allele dominant or recessive can also be affected by what you are measuring.

    In fact I wouldn’t be surprised if the same thing turns out to be the case with this argument about selfish genes. Just because an alteration in gene expression doesn’t seem to be accompanied by a change in DNA sequence only means that it wasn’t accompanied by a change in DNA sequence in the very small region of the genome in which you looked.

  31. 31
    Area Man

    I read PZ’s original take and this latest and agreed with it. Then I went and read Coyne’s take and agreed with that too. Then I went back and read Dobbs’s piece and found it mostly without merit. I think that’s the real problem. Dobbs bends over backwards to be contrarian and hypes ideas that are neither new nor earth-shattering. It’s like someone taught him biology in 1976 and then teleported him to 1990 so he could write about the amazing new things he learned. The article is bloated and yawn-inducing.

    To get to the point, a gene-centric view and appreciating the complexity of gene expression and phenotype are not mutually exclusive. You can go too far in either direction — being overly reductionist or ignoring what the molecules actually do, because complexity. But it’s not an either/or thing. And it’s a profound misunderstanding to believe it is.

  32. 32
    Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden

    @Chris61

    Okay, so, hypothetically, let’s say the DP gene creates a protein that, when it’s properly folded (suspend disbelief here) preferentially reflects different colors depending on orientation to incoming light.

    instead of a regulatory gene producing 2 different proteins, 1 protein, but 2 different colors.

    Now suppose that the protein is also used in digestion signaling. The DP protein is less sensitive than the BR protein in digestion signaling, so DP can turn on X pathway, but not Y, while BR is sufficiently sensitive to turn on both.

    Now we get blue light reflectivity when DP is present, but not when it’s not. Red light reflectivity is always present with BR or DP.

    Therefore, we get iris colors like above, DP appears dominant.

    BUT

    some gastroenterologist notices that that Y digestive function operates when BR is present, even if only 1 copy is present, but that Y digestive function does not operate when only DP is present. X digestive function operates for either version.

    Therefore we get BR as a dominant gene for digestive function.

    Same allele, same gene variants. One is “loss of function” compared to the other in color reflectivity, the other is “loss of function” compared to the other in activating signaling pathways.

    Both genes appear to be dominant, both appear to be recessive, but one in one context and one in another.
    ==========

    am I completely out to lunch?

  33. 33
    Kimpatsu

    …it’s activity is dependent…
    PZ, why have you suddenly gone over to the dark side and started misusing the apostrophe?
    And do you let our students get away with this butchering of grammar as well?

  34. 34
    chris61

    Genes aren’t dominant or recessive, it’s particular alleles of a single gene that are generally referred to by those terms. But certainly you can have alleles that will result in a relatively mild phenotype when present in a single copy and a much more severe phenotype when present in two. Depending upon how closely you look at the former you may not detect an abnormal phenotype at all, which would lead to calling the allele a recessive allele when if you look more carefully you’d consider it a dominant allele. Or in the case of human genetics, if you’re a cardiologist you might identify families with an autosomal recessive disorder (because only patients with two copies of the mutant allele have a cardiac problem) that a doctor looking at some other organ system (maybe a gastroenterologist) might classify as an autosomal dominant disorder (because one copy of the mutant allele causes a phenotype).

  35. 35
    johnharshman

    I’m talking about development and evolution. There certainly is far more to evolution than genomic sequences — those sequences need an environment (genes, cytoplasm, signals, etc) in which to operate.

    Sure, there’s a lot to evolution: all manner of environmental variables, from genetic context to cytoplasmic effects to other organisms to physical factors. But what evolves? What makes one species different from another species, or a current population different from its ancestors a million years ago? Changes in the genome. Period. All other changes — and that’s what evolution is — flow from that. Do you disagree? If not, then evolution, at bottom, is about genomes. And we can decompose those genomes into sequence chunks, or selfish genes if you prefer. Dobbs, from headline to major claims, is wrong.

    I took the cow statements as rhetorical flourish. But still, most of a cow is the same as a human and vice versa — it’s more than epigenetics, though, it’s changes in gene regulation that are inherited (and ok, some trans-acting changes).

    It’s changes on many levels, of which perhaps 90% have no effect on phenotype. But of that small percent that do, it’s everything from regulatory changes to protein sequences to a fair number of gene losses and novel genes. All of which are mutations to DNA sequences. And I think the cow statement, far from being a flourish, encapsulates Dobbs’ central point: the replacement in evolution of genes by epigenetics. Urk.

  36. 36
    Lenard Lindstrom

    Does gene driven evolution imply that the genome be the only mechanism of inheritance? For instance, there is this sub-population of humanity where the adult males lack a foreskin. Though random cases exist in the general population, foreskin absence predominantly runs in families. A man without a foreskin almost certainly has a father without a foreskin. So, for which mutation of what gene is circumcision a trait?

  37. 37
    Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden

    What makes one species different from another species, or a current population different from its ancestors a million years ago? Changes in the genome. Period. All other changes — and that’s what evolution is — flow from that. Do you disagree?

    I don’t know nearly enough to disagree but I can say that there’s a cytoplasmic environment “inherited” through the ovum. These “initial conditions” affect the next generation, but they aren’t genetic.

    Does that affect speciation – can changes in cytoplasmic initial conditions over the course of generations cause such a significantly different cytoplasmic environment that it could sufficiently change gene expression to merit labeling the new population a separate species?

    I can imagine this. I don’t know if it happens, but I know of nothing that says it is implausible.

    Of course, I’m not a biologist, so my ignorance is almost to be expected. I’m open to learning about the factors that limit variation in cytoplasmic initial conditions and/or the effects of same. But it’s weird to see people just assert that the only mechanism of inheritance is genetic.

    **obviously** there’s inheritance through cytoplasmic initial conditions. It might be insignificant to development, so there might be only one **significant** mechanism of inheritance. But there’s this other mechanism that does exist. Can you explain to me why I would be a crank to believe that the effects on gene expression caused by the cytoplasmic environment can cause differences sufficient to cause speciation?

  38. 38
    unclefrogy

    maybe because it is this time of year or its late who knows. As I was reading this and hearing about the complexity of the gene interacting with the other genes and their “products” I thought of food and the digestive system taking the results of gene interaction and breaking them down into smaller unites and the body > hence the cells taking all those smaller unites to do what ever they need to do, cause interactions and results influencing other results. The energy for all that chemical reaction and synthesis being derived from other reactions in plants and bacteria themselves made up of all that interaction and connected to it and all those atoms. None of it separate in isolation from any of the rest all having a different time frame, nothing stationary nor unmoving no place no thing. everything coming together for the discrete snapshot we take to facilitate our understanding of what are just events in time.
    uncle frogy

  39. 39
    grahamjones

    An example where a difference in developmental plasticity has evolutionary consequences: horses and zebras. Wild horses can be ridden. It’s not easy but it’s possible. Horses can be broken in – they are plastic. Not so zebras. No one has managed to domesticate them. As a consequence, over the last few thousand years, the horse population has expanded and the zebra population has declined. Gene frequencies within the horse population have shifted in quite different ways to what has happened among zebras. If things carry on this way, zebras will become extinct, and horses are likely to speciate.

    Of course it all comes down to genes eventually. Before anyone had ridden a horse, genes for being ridden on were selected because … well I don’t know why, but there must be a gene-centric explanation somehow. So gene-centric evolutionists can relax, secure in the knowledge that they still have a parsimonious explanation for everything.

    I am a mathematician who works with evolutionary biologists. It seems that a lot of evolutionary biologists have a bit of a fetish about parsimony. (I’m not referring to my colleagues!) But scientific models are judged by their utility not their parsimony (as wolfgang says at #16). The acid test is whether, when you want to make a prediction about evolution, you find it beneficial to incorporate developmental plasticity into your model. For example, we would like to predict which bacteria will evolve resistance (that is, acquire genetic mutations which confer resistance) to which antibiotics. Should the differences in developmental plasticity between different bacteria be taken into account when making such evolutionary predictions?

  40. 40
    johnharshman

    Can you explain to me why I would be a crank to believe that the effects on gene expression caused by the cytoplasmic environment can cause differences sufficient to cause speciation?

    Yes. The cytoplasmic environment you’re talking about here consists of maternal proteins and transcripts. Differences among species in their sequences and concentrations within the cell are both determined by genetic differences.

  41. 41
    jamessweet

    This is somewhat tangential, but one thing that bugged me about Dobbs’ article is that he kept asserting that because we share so many genes with other animals (“We are 60% cow!”) that there must be more to genes. I don’t think that necessarily follows… there is so much complexity in just building a basic cell. There is still incredible complexity (though somewhat less so, I think) in differentiating types of cells. How those differentiated cells are actually laid out in different organisms? Well, that’s still pretty complex, but now it’s starting to seem more manageable.

    Two Minecraft worlds might be entirely different, but most of the bytes needed to make that world appear on your computer are contained in the Minecraft program itself — the code that tells how each square interacts with the others, etc. The data file with the actual contents of the world, while crucially important in determining how each world appears, is small in comparison. It doesn’t seem at all strange to me that the same might be true with genes.

    To be clear, this doesn’t refute anything Dobbs’ says… I just felt like that repeated reminders of how much DNA we share didn’t really score him any points, at least not with me.

  42. 42
    vireyda54

    I read the article on PZ’s initial recommendation because while I play with rocks all day, I like to read casually about evolution and read TSG many years ago and enjoyed it. (the most I’ve ever been taught about the subject was in a paleontology class quite a while ago)

    I didn’t find David Dobbs’s article nearly as compelling as this response. The Aeon article had personal attacks and what seemed like hyped up controversy. I would have enjoyed his article and understood he wasn’t completely shouting for paradigm shift if not for those elements.

    I vote next time he wants to tackle the subject, he ought to ask PZ to write the thing instead.

  43. 43
    vireyda54

    @ Area Man

    That was really how I ended up reading everything too.

    Also.. shame on me. Dobbs’s should be Dobbs’.

  44. 44
    ChasCPeterson

    it’s weird to see people just assert that the only mechanism of inheritance is genetic. **obviously** there’s inheritance through cytoplasmic initial conditions.

    you should have retained the scare-quotes. That’s not inheritance. It’s a phenotypic trait of the mother organism, being present in her reproductive cells before fertilization, and it’s genetically based.

    It might be insignificant to development, so there might be only one **significant** mechanism of inheritance.

    It’s certainly not insignificant to development, being as how it runs the show for at least the first several cell divisions, but it’s not a mode of inheritance separate from the genome (as pointed out by johnharshman already).

    Can you explain to me why I would be a crank to believe that the effects on gene expression caused by the cytoplasmic environment can cause differences sufficient to cause speciation?

    Because the only way such effects could get passed on to another generation would be for them to be genetic. One generation does not a species make.
    (More pissily, you could hypothesize such a scenario without being too crankish, but if you believe it without evidence, you’re unclear on the whole concept.)

    Dobbs’s should be Dobbs’.

    nah. “Dobbs’s” is fine. I actually prefer it, so when MY stylebook becomes Law, it will be the Only Right Way!!!!!

  45. 45
    ym su

    Some metaphors and/or examples…

    -the relationship of gene to phenotype is like God conferring free to man (by the way, I was raised atheist and have no religious background so I may be wrong… but God is fiction so there’s no right).

    -heritability of culture

    example: compare phenotype of cultures which promote vaccination to those which do not

    example: compare jaw morphology of persons in hunter-gatherer cultures to those of the developed world

  46. 46
    Vilém Saptar

    PZ :

    I will be told again that those processes are ultimately a product of genes, and I will say in vain…but you don’t account for all the cellular and environmental events with sequence information!

    Point taken.

    We wouldn’t argue that if we understood every atom in the human body, we’d understand poetry, after all.

    Of course we wouldn’t. But, I’m not sure that’s applicable here. Or your computer analogy. We aren’t looking at a different level of description. We’re just looking at the same level of description with factors we hadn’t considered before. Factors that are a product of the historic trajectory of some elements ultimately governed by genes with a good amount of randomness thrown in. I know you already pre-empted this objection, but this does not sound like a different level of description to me at all. So I’m not sure I’d be onboard with that objection. To my non-biologist ears, epigenetics seems to be another force at the same level. And largely determined by genes. Sorry to be going round in circles.

    As for Dobbs’ piece, Chas is right. It was more in the vein of “zOMG Phenotypic Plasticity!1!! Dawkins’ evil influence on evolution is dead zOMG”, than your more nuanced position.

  47. 47
    chrislawson

    PZ, to me the major problem here is not that you’re arguing that we need to look at more than just genes to understand evolution, but that you’ve hitched this particular wagon to an article by Dobbs that is poorly written, confused, hyperbolic, misleading and now has the author using weasel words to pretend he didn’t say what he was clearly saying.

  48. 48
    davidwhitlock

    The problem is human hyperactive agency detection. A flaw of human cognition is to impute an “agent” that “causes” things to happen and be the way that they are. This is a “feature” (bug) of predator detection. False positives (running away) are lower cost than false negatives (being eaten), so evolution has skewed sensory and data analysis systems with a hyperactive agency detection bias. We know this intellectually, but because our whole nervous system is wired this way, this way of thinking is very compelling. Our brains evolved for us to survive and reproduce, they did not evolve to understand science.

    It is this hyperactive agency detection bias that causes humans to impute agency to genes and use a gene-centric model. There are simple examples where a gene-centric approach seems to “work”. It doesn’t work in all cases, and can’t work in all cases because it is a fundamentally flawed approach. What is flawed is the “centric” approach. There is no “center”.

    People really, really, really, really want there to be something that makes them “special” and “better” than their friends and neighbors, and especially somethings that makes them and their friends and neighbors “better” than people who look differently, people who have different colored skin, eyes, hair, and different shaped bodies and who speak different languages. People need something to justify feeling “special” and “better” so that they can justify treating those “others” badly.

    Physiology comprises at least hundreds of thousands of coupled non-linear pathways. We know that systems of coupled non-linear pathways become intractable with n greater than 3. There is no “center” in a system of hundreds of thousands of coupled non-linear pathways.

    Applying a linear scale to such a system may be “correct” in a mathematical sense. Cows and humans may share 60% of genes or genomes, or whatever. Cows and humans share virtually 100% of chemical elements. Mostly hydrogen, then oxygen, then carbon, nitrogen and some others. So what? The degrees of freedom are so large they are difficult to estimate. 10^5 factorial? Per cell?

    The phenotype is not specified by the genome in the sense that the “gene-centric” view imputes. The number of heads a person has is not specified by the genome. The number of heads a phenotype has depends on the minutia details of development at the earliest stages of development. For example this phenotype.

    http://en.wikipedia.org/wiki/Abigail_and_Brittany_Hensel

    If there had been a little more adhesion between cells at a certain stage of life, the phenotype that resulted would have one head. A little less adhesion and they would be MZ twins.

    There isn’t a one-to-one correspondence between genotype and phenotype. A single genotype can produce astronomical numbers of different phenotypes through developmental plasticity. Yes, that “plasticity” is mediated through the genome, but the “degrees of freedom” of the plasticity exceeds the “degrees of freedom” of particular genes.

    The idea of genes following phenotype changes is quite compelling, but is anathema to a gene-centric view. A gene mutation can’t produce such a large phenotype change that speciation occurs in that single generation. If it did, then the organism expressing the mutant gene would die without offspring (because it is a different species and has no potential mates). Mutations need to be compensated for by developmental plasticity until there is a “need” for the phenotype differences that the mutant gene(s) accommodates at reduced cost. Having big muscles with less exercise is good, if you need big muscles. If you don’t need big muscles, it is better for them to atrophy so you don’t have to expend the calories carrying them around, repairing them and keeping them healthy. You can use those calories for more descendants. Atrophy of muscles through disuse is a “feature”.

  49. 49
    Seize

    Whitlock, you just lost me at paragraph 1. You fell into an ev-psych pothole. Humans are not predator-detectors; we are human-detectors. Many animals in our clade are far better predator detectors than we; we’re most skilled at evaluating the potential actions of our fellow humans using various sorts of cognitive models. In such a schematic of two opposed, equal humans with competing interests, a false-positive (suspecting another human wants your mate) might be as costly as a false-negative (supposing they don’t).

  50. 50
    Claudio

    Of course epigenetics matter, even for evolution and I don’t think Jerry/Richard think it doesn’t. Jerry is just reacting to the postmodernist claims that epigenetics will ‘re-write’ biology and all that bull crap. C’mon everyone what’s going on?

  51. 51
    A. R

    I think a quote from a recent student essay would be appropriate:

    “Genes, as major contributors to organism morphology must be considered necessary, but not sufficient for evolution by natural selection. The interplay of genetics, development, and environmental influences on development must not be ignored as major actors in the response to selective forces.”

  52. 52
    Alf Hickey

    PZ – “I’m talking about development and evolution. There certainly is far more to evolution than genomic sequences — those sequences need an environment (genes, cytoplasm, signals, etc) in which to operate”

    Cite please? I would love to see an example of non gene centered evolution.

  53. 53
    ryancunningham

    @52 Alf Hickey

    “Cite please? I would love to see an example of non gene centered evolution.”

    http://en.wikipedia.org/wiki/Cis-regulatory_module

    http://en.wikipedia.org/wiki/Histone_code

  54. 54
    Max

    I feel like PZ is having a completely different argument from everyone else. None of what he posted contradicts or rejects the gene-centric view of evolution. That environment plays a role is a huge DUH that everyone already knew about even way back in 1976. Yes, it’s awfully complex, and I’m glad there are people like PZ out there devoting their careers to studying it. But Coyne is 100% right that there’s no paradigm shift here. As fascinating and confusing and cool as all these discoveries are, they’re just small tweaks to our existing model.

  55. 55
    davidwhitlock

    Seize, the first type of agent detector was (very likely) a predator detector. Very likely human ancestors had them when those human ancestors were still single cell organisms. When human ancestors became multi-cellular and required more complicated predator detectors, human ancestors evolved them by elaborating on the already existing predator detection systems.

    Fast forward 500 million years, and for virtually all of that time, predator detection was a key use of those “detection systems”. The false-positive bias has been built into those systems for at least 500 million years. This is why it is difficult to be a scientist, and why (as Feynman says) “you are the easiest person to fool”. It is our evolved hyperactive agency detection system that makes us the easiest person to fool.

    It is hyperactive agency detection that compels people to impute that there must be a “center of evolution”, so that evolution can be “gene-centric”. It is more complicated than gene-centric.

  56. 56
    A Masked Avenger

    ChasCPeterson, #44:

    Almost all of the science in this thread is going over my head: I’m a mathematician with a high-school knowledge of biology, where “high school” was in the Reagan era. That said, I think this comment misses something that I was going to mention:

    Because the only way such effects could get passed on to another generation would be for them to be genetic. One generation does not a species make.

    If the expression of one’s genes is influenced by gestational conditions, then I fully get that the next generation will not be directly affected by its parents’ gestational conditions (although it’s unclear to me that this could never be possible); but it will be influenced by its own gestational conditions. Specifically, a clone can be expected to be more different from its original than a twin would be. When reading about the cloned cat, I realized that “identical twins” are not only genetically identical, but also shared more or less identical prenatal conditions.

    The upshot is that this suggests to me, qualitatively, that a complete knowledge of my genome is insufficient to make detailed predictions about me. You also need to know all about my mother’s genome, in order to deduce some things about the conditions under which I developed–and not only that, but you’ll need to know incidental facts about my gestation, such as any illnesses my mother might have experienced, or nutritional deficits, or her level of physical activity, or any medications she may have been on. Most, though not all, of these things are genetically determined, but they’re determined by the genes of my mother, any bacteria and viruses she was exposed to, etc.

    The upshot of that is to hypothesize that if you had split my zygote numerous times, and implanted all my twins in different mothers, then you would see some variation among us, despite our identical genes. If you wanted to analyze the effect of my genes, you could then not say that they code for an eventual height of 6’3″, or a single IQ or hair color. You’d be forced to say that among my twins, the average height is such-and-such, which may be a few inches taller than the average height of the general population, and so on.

    This might be such common knowledge to you biologists as to be trite, so I submit it hesitantly. Discussions like the one PZ describes don’t sound as if they include a recognition that many, perhaps most, phenotypic traits are such an indirect result of genetics that you can’t properly speak of my genes doing more than determining a certain probability distribution for this or that trait, which differs from the probability distribution determined by your genes.

    If I’m reading PZ with any comprehension at all, then I think my previous paragraph would be an acceptable summary, stripped down to barest layman’s terms.

  57. 57
    patrick85

    @davidwhitlock

    To echo your point, some commentary from Alan Watts (a bit cobbled together, and worth quoting at length):

    “Generally speaking, we have two kinds of consciousness. One I will call the ‘spotlight,’ and the other the ‘floodlight.’ The spotlight is what we call conscious attention, and we are trained from childhood that it is the most valuable form of perception. When the teacher in class says ‘Pay attention!’ everybody stares, and looks right at the teacher. That is spotlight consciousness; fixing your mind on one thing at a time. You concentrate, and even though you may not be able to have a very long attention span, nevertheless you use your spotlight: one thing after another, one thing after another…

    However we also have floodlight consciousness. For example, you can drive your car for several miles with a friend sitting next to you, and be completely absorbed in talking to your friend. Nevertheless, your floodlight consciousness will manage the driving of the car, will notice all the stoplights, the other idiots on the road, and so on, and you will get there safely without even thinking about it.

    It’d be just too much to think about with a single-pointed ordinary attention consciousness, which is a scanning system, like radar.

    So ecology is man’s relationship to the world, or a plant’s relationship to its environment. All that kind of relationship, the study of the bee and flower bit, is ecology.

    The thing that is so characteristic, then, of this new or different kind of consciousness, is that it starts from or has its foundation in awareness of relationship, of ‘go withness,’ that the inside of a situation goes with the outside, and although you may think from the point of view of ordinary consciousness, that they work independently from each other; in this state of consciousness you see that they don’t. In other words, it’s slowly beginning to penetrate our ordinary consciousness. That what any individual does, and we ascribe to him as his behavior and praise him for it or blame him for it, everything that he does goes with what happens outside him. The behavior of the environment, and the behavior of that organism within that environment, is one behavior, and you mustn’t think of this deterministically. That is to say, as if the organism were something merely subservient to the environment. Nor must you think the opposite way, that the environment is something that can be pushed around by the organism. When an organism starts looking as if it were pushing its environment around, it simply means that the environment/organism, the total field, is changing itself.

    So there is no determinism in this, just as there is no idea of old-fashioned free will. You learn to see that there is simply one behavior pattern working, which we will call the organism-environment, and if you understand that, you understand that YOU are this totality organism-environment, and so you are moving with it in the same way that all the organs of your physical body are moving together. As all the cells of the brain cooperate. You don’t have to make them cooperate, you don’t have to tell them to; you don’t have to arrange a treaty of some kind, they just do so. So when birds fly, you notice particularly birds like sandpipers, when they turn suddenly in the air, they turn as if they were all one bird. Although when they land on the sand, they become individuals, and they run about independently looking for worms. Then suddenly you shout at them, and they shoot into the air, and they’re all one creature, moving as if it had a single mind. You know that haiku poem:

    A hundred goods from the mind of one vine.

    So just as we are organized that way, as organisms, so also we are, although not aware of it, organized that way collectively as individuals relating to each other and relating to the other forms of life, and to the geology, and the meteorological and astronomical phenomena around us. Only we haven’t come to notice it. Our attention has been so fixed upon some of the details of this relationship, that we have created a system of details as if it were a separate physical system. You understand, I’ve mentioned this, I’m sure, to many of you before, that human beings have for at least 3000 years specialized in one kind of attention only. That is what we call conscious attention, and that is a form of scanning the physical environment as if we were looking at it with a spotlight. And therefore, the nature of scanning is this: that it takes in the whole scene in series, bit by bit. Even if you don’t go in a straight line, and you scan looking around you, you have a series of glimpses or glances piled up, and that gives you the history, in linear time, of your existence, because it’s one experience of attention after another.

    We screen out by the way we pay attention to things. We are absolutely befuddled with words, and you see, words follow the same linear pattern, because words are a notation. Conscious observation of the world by the spotlight always is accompanied by a notation. That is to say the notation of language, the notation of written letters, the notation of numbers, the notation of algebraical symbols, any kind of notation you want to think of. Musical notes–they do the same thing. And you notice what you can notate, and that is what is notable, noteworthy, because we observe and become aware consciously only of those things that we consider important. And what do you consider important? Well, that depends on your hobby. For which for most people is survival.

    [It] is a result of overspecialization in linear consciousness. Now, linear consciousness is indeed remarkable, but it is something in a way aggressive. Just as the sword, the cutting edge, is an aggressive instrument, as distinct from the total skin. With the total skin, you can feel all over, and in this way you embrace life. When you get into a hot tub, it goes all over your skin, and it’s a type of diffused thing, what Freud called polymorphus erotic feeling, all over. Whereas conscious awareness is like the point of a pencil: it jabs, and it writes down precisely what. And so those people who are all conscious attention are sort of intellectual porcupines. They’re all prickles into things, and that gives them an essentially hostile attitude toward life, because of course conscious attention is a troubleshooter. It’s the radar in the human organism to watch out for changes in the environment, just as the radar of a ship is watching out for icebergs, and an airplane’s radar is watching out for thunderclouds. So in the same way, our thing is going around like this, and it’s serving a very valuable function. But if you identify yourself all entire with that part function, then you define yourself as being in trouble, and looking for trouble, and you become unaware of your generalized relationship with the external world.”

  58. 58
    Alf Hickey

    @ 53

    Can you point out an example of when the Histone code or the Cis-regulatory module have been at the center of the evolution of a new trait through natural selection?

  59. 59
    ChasCPeterson

    Masked Avenger: No argument with anything you said. You’re talking about what biologists refer to as ‘maternal effects’, which can include things like where a turtle lays her eggs (the same egg in a shaded nest is more likely to hatch a male than it would in a sunny spot), what hormones and how much are present in the egg yolk (birds and reptiles) or circulation (mammals), etc. This is an active area of research because there is no doubt that such maternal effects can profoundly change phenotypes of offspring. You’re correct: phenotypes cannot be 100% predicted from genotypes alone, even in theory.
    But the thing is, either these maternal effects are themselves genetically based or they’re not, and if they’re not, they can’t evolve.

  60. 60
    Ryan Cunningham

    @58 I could, but I’ve done enough research for dishonest goalpost movers in my lifetime. If you don’t think natural selection acts on cis-regulatory elements, you’re beyond a crank.

  61. 61
    johnharshman

    Of course natural selection acts on cis-regulatory elements. They’re DNA sequences. So that isn’t any sort of challenge to the selfish gene metaphor or any support for the evolutionary importance of non-genetic factors. And histone modifications aren’t inherited (or not much) and can’t be important in evolution either. So while Alf was wrong about the nature of cis-elements, he was right that neither of those is an example of non-gene-centered evolution.

  62. 62
    Alf Hickey

    @60

    Sorry, that wasn’t want I meant. I was probably less than clear in my post.

    johnharshman posted (and I agree with him),

    “PZ, you seem to be thinking about development while Jerry is thinking about evolution. All these epigenetic effects are important in development, but they aren’t what changes in evolution; that’s all genomic sequences.”

    PZ responds to this by saying,

    “I’m talking about development and evolution. There certainly is far more to evolution than genomic sequences — those sequences need an environment (genes, cytoplasm, signals, etc) in which to operate.”

    In his article Dobbs seems to claim that epigentic and developmental effects are equally (and maybe more) important than genetics for evolution. And PZ is saying he agrees. I’m asking for some evidence of this. I don’t think there is any.

    (As an aside, who would ever claim that all evolution needs is genomic sequences? This is a total straw-man).

  63. 63
    anchor

    From my reading of this since it started with Dobbs’ article: I did NOT like it: his gratuitous snipe at Dawkins had absolutely no legitimate place in an article attempting to pose a reconsideration of a scientific metaphor or paradigm.

    What the hell are we to make of a journalist who uses such a device in this context? That Dawkins is ill-equipped to write on the subject because of some perceived ignominy in his twitter postings? Thats like punching out with an OT valve before the get go, and its a terrible and hypocritical front on which to carry the proposal suggested by the article that supposedly aims to refute those ideas on a legitimately scientific basis.

    But it doesn’t stop there. In his effort to make his case, Dobbs characterizes Dawkins’ position so inaccurately that I am convinced he hasn’t understood anything in Dawkins’ “The Selfish Gene”. As many others have already surmised and noted (curiously, more often and a great deal more clearly over at Jerry Coyne’s site) it seems he has glued his argument against the impression supplied by the TITLE of that book, rather than upon a careful reading of its contents.

    That charge is quite compelling from all of the comments I’ve read on all of the relevant sites to this point.

    Max #54 is right. And so is chrislawson #47.

    Jerry Coyne supplies, in two parts, typically excellent and compelling arguments, which in this case demolishes Dobbs’ easy dismissal of Dawkins’ thinking. Coyne sees immediately that Dobbs’s article FIRST places a premium on the notion that Dawkins’ thinking was ever at odds with the consensus view, or that such a consensus was ever in conflict with Dawkins’ ideas. Dawkins, as a legitimate scientist in that field, legitimately contributed to that field and his ideas have been roundly accepted in that field. Not that ANY who accepted those ideas EVER thought that the question of the complexity of evolution through the many avenues of selection or its breakage (both macroscopically and phenotypically environmental as well as microscopically and genotypically environmental – whatever inevitably goes on upon a THERMODODYNAMIC/CHEMICAL/MICROENVIRNOMENTAL playing field (pardon my guesses on how best to cover the bases in terms of proper terminology) which may indeed determine what happens not only on a statistical chemical basis, but on the whim of, say, sudden exposure or immersion in a very different extra-cellular macroscopic chemical environment or, with the passage through a cell containing such relevantly important genetic information in the form of molecular structures, of a single sufficiently high-energy ionizing particle courtesy of the ubiquitous and constant rain of cosmic rays life is exposed to…that big picture, which I suspect is what PZ is addressing, is definitely on the table.

    I don’t understand how this is disputable, or that anybody gets their hairs curled over the notion that this in any way defrocks the role of genetic inheritance as it developmentally flowers to phenotypic expression as so excellently understood thus far!!!

    Yet we see idiotic comments on Dobbs’ site from the inevitable fanboy mentalities that suggest that Dawkins is somehow a false scientist and even that he has not supplied any mathematics to support his claims! And none of that crap has been corrected (as of my last checking) by Dobbs himself.

    That’s pretty low if its a conscious decision not to correct it…

    What’s even lower, however, is that Dobbs saw fit to buttress his position by suggesting that Coyne’s rebuttal “willfully misreads” as well as “rudely and inaccurately” characterizes what he wrote, while tenderly submitting the camouflage of an admission of a “combative tone” on his part for which he pretends to apologize. And then he completely walks right off the end of the pier by complaining that he never ONCE mentions Lamarckism for which Coyne takes him to task…although it is abundantly evident that he describes exactly that in his treatise. He doesn’t acknowledge that his article words it very precisely thus. (In this, I think PZ is also mistaken: Coyne calls DOBBS out on HIS describing Lamarckism, NOT on any such tendency in what PZ says). Instead, Dobbs appeals to the excuse that it was not what he meant – although he is adroit enough to refrain from the specific occurrence – or maybe he simply doesn’t understand that what he wrote in those passages describes Lamarckism to a tee. Methinks its the former, but its hard to tell given how obtuse Dobbs is in owning up to any real humility he needs to be able to qualify as a legitimate voice on this dispute he has managed to conjure…yet, sadly, not without help by some of the experts themselves.

    BUT. Nevertheless. PZ is ALSO perfectly correct to point out (as he has a number times previously) that there are must be mechanisms at work which almost everyone also agrees must impinge upon evolution via the current canon of natural selection (which I am sure Coyne and Dawkins would agree) which must happen besides and addition to our strong yet inevitably sketchy understanding or principle-based models of how selection, ‘natural’ or otherwise, might operate in the actual world, as full of complexities as it is, and through such acknowledged complexity, we by no means have yet adduced in its complete glory- a question Dobbs, in my opinion, is clearly incapable of addressing.

    Not if he’s miserable enough to couch his premise upon a perceived condition of Dawkins’ character based on tweets, no matter how outrageous they may have been. WTF???

    I think PZ made a mistake of precipitously hanging his point onto Dobbs’ article without suitable care in reading what was actually there in it. It is clear to me from my studying the whole and the associated commentary it did NOT in any particular way support anything PZ had previously said on the subject. And from what PZ has exhaustively emphasized, I’m sure he doesn’t subscribe to any notions that genes are not ostensibly responsible for preserving phonotypical traits, precisely as Dawkins has repeatedly emphasized it, by cumulative selection upon an established history of selected genetic configurations that express themselves as traits.

    We can and do all make mistakes, especially when we see anything that appears to agree with what we have been trying to say. Digestible appearance makes it easier to rush to a statement of support.

    On this it is Dobbs who holds the burden, not Dawkins, not Coyne, not even PZ.

    What’s so goddamned troubling, though, is that whenever anyone finds flaw in another (or their honest scientific arguments) however slight, it seems, the inevitable response is to hunker down and defend, rather than open up and allow that one may have been either wrong or mistaken in accepting another’s take that doesn’t hold water.

    Its doubly troubling that a ‘journalist’ then can get scientists in such an uproar – complete with – and this is the worst aspect – cheering sections, as if its a goddamned football game. Sheesh.

    Coyne is right. Dawkins remains intact. PZ is also right. Dobbs was in way over his expertise, and on top of that he showed himself to be under any standard of proper scientific discourse, and he ought to be dismissed from the discourse, period. In the meantime, it would behoove all to keep their steady aim upon what they all hold within their understanding WITHOUT resorting to a sniper position – as if opinion ever settled a genuine scientific controversy, let alone a false controversy bubbled up by an ambitious upstart such as some guy who doesn’t matter, EXCEPT (and this is the REAL problem) how these guys constantly publicly promote the idea that a scientific consensus reflects an uncertainty that people should find as a reason to distrust what scientists say. Its despicable.

  64. 64
    Ryan Cunningham

    @62

    You don’t think there is any, but there is. Look up crm studies in drosophila. There are lots of examples of regulatory sequence being removed or modified that changed the development of the organism. If you don’t care to read about them, fine. That’s your choice. You can continue to believe in the oversimplified selfish gene model that initiated this entire controversy.

  65. 65
    gillt

    @61: johnharshman

    And histone modifications aren’t inherited (or not much) and can’t be important in evolution either.

    It’s not clear what you’re saying. Obviously chromatin is dynamic, but non-protein-coding stretches of DNA that modify histones are conserved in mammals. Less than protein coding regions, but still.

    Look at the literature on “promoter-associated long non-coding RNA elements” and see if you agree.

  66. 66
    Marc Abian

    Crip Dyke #3

    I still think it was off base – when I read it, it didn’t seem Lamarckian at all (parents sprint more b/c they have to even though they aren’t in shape for it, kids sprint more b/c following parents and so are in shape for it before they get to adulthood and are better at it, grandkids sprint more b/c they are following parents who were better at it than the grandparents were, thus same dynamic, but with better teachers).

    In what sense have these faster offspring evolved to be faster? The slower adults had the capacity to run just as fast, but they didn’t have the need or the better teachers. The way I read it Dobbs thinks a plastic response is locked in by a gene mutation (gene doesn’t lead, it follows), so the offspring must get faster through training, and only then can a mutation take place which allows this phenotype to become cemented in the gene pool.

    #64

    There are lots of examples of regulatory sequence being removed or modified that changed the development of the organism.

    Is that change genetic, epigenetic or developmental?

  67. 67
    johnharshman

    Modification of histones isn’t inherited. Stretches of DNA that modify histones are inherited. That’s genetic. Removal and/or modification of regulatory sequences is genetic too. What differs among species that isn’t either DNA or a product of DNA?

  68. 68
    vireyda54

    @64 anchor

    I was a bit annoyed at Dobbs’ response article where I made a comment answering his question of why someone might have misunderstood his new stated purpose because of personal attacks and over-hype and then it was monitored and not posted.

    I think Dobbs wanted the attention, but like you said, can’t really handle the criticism. PZ has made a compelling, well reasoned case for broadening our understanding of how things work. I don’t think he needs Dobbs’ help.

  69. 69
    gillt

    @67 Johnharshman

    Modification of histones isn’t inherited. Stretches of DNA that modify histones are inherited.

    Obviously proteins aren’t inherited. So what’s the difference? What’s the difference between an inherited DNA element that modifies a histone (or more DNA) and the predictable modification of a histone that’s a result of said inherited DNA element? Why can’t we “model”–in a similar way we do with alleles–methylation profile patterns or isochores in populations…or am I missing something?

    What differs among species that isn’t either DNA or a product of DNA?

    And the way you worded your question makes the plain as day answer uninteresting and trivial. Which if I hazarded a guess was the point of the OP and where Myers and Coyne part ways.

  70. 70
    David Marjanović

    An example of a “selfish gene” is a Homing Endonuclease Gene.

    *sigh* That’s not what Dawkins means! This is only one way a gene can be selfish! Every gene is selfish in the way Dawkins means!

    Dawkins hypothesized that if a gene resulted in an association of light skin and behaviors favoring others with light skin, that such a gene could become dominant in a population. This is not the case because such a gene would be out competed by a gene that simply provided light skin without any of the favoring behaviors (which impose a cost).

    You assume that it’s reasonably easy for that other gene to evolve. It’s not clear how often this is the case, but it’s obviously not always: otherwise we’d all be merrily making cellulase.

    PZ, why have you suddenly gone over to the dark side and started misusing the apostrophe?

    “Suddenly”? PZ has never used the spelling its as far as I’m aware, which covers several years on Pharyngula.

    And do you let our students get away with this butchering of grammar as well?

    Spelling isn’t grammar. And grammar still wouldn’t be the topic of whatever the students write.

    In such a schematic of two opposed, equal humans with competing interests, a false-positive (suspecting another human wants your mate) might be as costly as a false-negative (supposing they don’t).

    Hardly. Being nervous around that person will far less often decrease your evolutionary fitness than a false negative will.

    The way I read it Dobbs thinks a plastic response is locked in by a gene mutation (gene doesn’t lead, it follows), so the offspring must get faster through training, and only then can a mutation take place which allows this phenotype to become cemented in the gene pool.

    That’s an old idea, and nonsense if taken as literally as you just spelled out. If such a mutation happens after all this, then it can be selected for; if it happens earlier, it’s useless at best and won’t spread through the population. Mutations happen at random; they are not caused or made possible by selection pressures.

  71. 71
    johnharshman

    Why can’t we “model”–in a similar way we do with alleles–methylation profile patterns or isochores in populations…or am I missing something?

    Let me put it another way. When we’re talking about evolution, it’s all about descent with modification. What descends, i.e. what is inherited? Genotype. Methylation profile patterns and isochores can be considered phenotype. Phenotype is of course what’s directly subject to selection, but it’s the underlying genotype (or rather the distribution of genotypes in the population) that is the effect of selection and is what makes selection meaningful over evolutionary time. So of course you can model phenotype and even trace it over time. But in evolution, it’s genotype that we need to be looking at.

  72. 72
    gillt

    @71 johnharshman:
    I’m understanding you less now. Alleles are the only thing that descends with modification is what you’re saying. Other genomic features such as long stretches of coding and non-coding DNA (isochores) don’t count and evolutionary changes to chromosomal number driving diversification of a species is a misguided hypothesis.

  73. 73
    johnharshman

    You are right that you’re understanding me less. If by alleles you mean versions of protein-coding sequences, which is all I can understand you as meaning, then I have never said or, I hope, implied such a thing. I used the term “genotype” to avoid all possibility of such confusion, or at least I thought so. Long stretches of DNA, coding and/or non-coding, are sequences. Changes to them are mutations. Of course they count. Changes in chromosome number are also mutations and changes to sequences, just the same as are indels. How is any of this pointing to non-genetic changes in evolution?

  74. 74
    Amphiox

    I don’t think we can assert that ONLY the genotype is heritable. Aspects of phenotype may be heritable, too.

    For the most basic example, consider the concentration gradients within the zygote that determine the first differential gene expressions in development. Where did that information come from? Nowhere in the embryo’s genome is that encoded. Was it encoded then in the mother’s genome. But she was once a zygote too, and it was not encoded in her genome then, and her genome has not changed barring somatic mutations since. It is a cellular phenotype, and aspect of which is not part of the genome, that has been inherited as cells divide their cytoplasm, in an unbroken chain all the way back to the first protocell sprung forth from abiogenetic processes. Never has it been encoded, not in its entirety, in any genomic molecule.

  75. 75
    ChasCPeterson

    Was it encoded then in the mother’s genome[?] But she was once a zygote too, and it was not encoded in her genome then

    I suspect it was, but it wasn’t being expressed at the time. If such gradients exist in ova before fertilization (as I think they do), then it is a maternal phenotype. If the gradient was not present in the oogonium or undeveloped oocyte (as I think it’s not), then it must trace back to gene epression.

  76. 76
    Alf Hickey

    How would an aspect of phenotype be heritable in a way that would make it available to evolution by natural selection?

    Concentration gradients within the zygote are biological forms that arise from the underlying physical structure of organisms. It is like saying that the spherical shape of a cell is “inherited”, when it really just arises due to the basic nature of membrane enclosed structures.

  77. 77
    Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden

    @mark Abian, responding to me:

    First me:

    I still think it was off base – when I read it, it didn’t seem Lamarckian at all (parents sprint more b/c they have to even though they aren’t in shape for it, kids sprint more b/c following parents and so are in shape for it before they get to adulthood and are better at it, grandkids sprint more b/c they are following parents who were better at it than the grandparents were, thus same dynamic, but with better teachers).

    Now Mark:
    In what sense have these faster offspring evolved to be faster? The slower adults had the capacity to run just as fast, but they didn’t have the need or the better teachers.

    They haven’t. That’s the point. Every organism must have some in-built flexibility. A certain amount of change in response to environment is vital, or we’d be plants. Hell, even plants need to be able to change in response to environment.

    The way I read it Dobbs thinks a plastic response is locked in by a gene mutation (gene doesn’t lead, it follows), so the offspring must get faster through training, and only then can a mutation take place which allows this phenotype to become cemented in the gene pool.

    In the new environment where selective pressure favors open running, a mutation may come along which provides benefits for organisms faced with that environment. But the response of animal plasticity is faster than the response of a gene dispersion through a population.

    The gene follows, it doesn’t lead. That doesn’t mean that the hypothetical running was “inherited” (although it might be epigenetically…but then I don’t know much about epigenetics or the power or persistence of epigenetic effects, so I’m not alleging any need for or even presence of such, just not ruling such out) from the original parents (now grandparents).

    Plasticity is a faster response to environment than gene distribution, and although genetic changes may ultimately be more powerful (especially because, e.g., there’s no reason you can’t get the benefit of greater running practice **and** your newly longer -thanks to genetics- legs) than plasticity in changing capacities of a species, an environment which selects for genetic traits that enhance running will of necessity be an environment that encourages plastic development towards better running.

    Isn’t this obvious?

    I’m not qualified to speak about whether it accomplishes “speciation” sometimes, rarely, or never though it plausibly could: if a population deep in the forest has no pressure to engage in open running, the forest-edge population subject to the fire now runs to catch pray…and thus feels confined by the forest, thus doesn’t venture into the forest as often, thus “causes” the speciation (though the speciation wouldn’t be considered accomplished until the populations were genetically distinct, the particular genetic distinctions would not, in this scenario, have caused the separation of the breeding populations).

    Nonetheless, the broad outlines of what’s being said and how it is non-Lamarkian seem so obvious that I would wonder if I was missing something if not for PZ apparently reading it the same way.

    Plasticity will lead genetic diffusion in adapting to new environmental conditions whenever it is possible for plasticity to lead genes in adapting to new environmental conditions merely because plasticity can operate on every individual member of a species at once and immediately. Genetic diffusion cannot.

  78. 78
    Alf Hickey

    Another way to think about this subject it to ask what is the difference between biological evolution and geology?

    Both fields consist of the study of change over long periods of time, which is governed by certain rules. Different rock types and various geographical features are analogous to the phenotype of an organism.

    What separates the two fields is that change in biological evolution is causally driven by the natural selection, while there is no such process in geology. And what allows natural selection to happen is a reproducible genotype.

  79. 79
    Marc Abian

    Crip Dyke #77

    They haven’t. That’s the point. Every organism must have some in-built flexibility. A certain amount of change in response to environment is vital

    Is that the whole point of his article? That’s all obvious stuff. How it could possibly seen as even worth an article, let alone as a challenge to Dawkins? I’m not sure it’s even a challenge to Darwin.

  80. 80
    gillt

    I never intended to make an argument about non-genetic changes driving evolution. I understand genotype to be a comparison of alleles (never mind coding or non-coding) between two or more individuals. Practically speaking, in phylogenetics this means making gene trees of SNPs but in the age of whole genome sequencing I see no reason not to accommodate isochores and other “higher order” genomic patterns, which you seem not to agree with, apparently.

  81. 81
    Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden

    @Marc Abian:

    I don’t think that’s the whole point of the article, but that was the whole point of that example.

    He uses the grasshopper/locust distinction to talk about how the same genes in different environments cause such different expression that we give the creatures different names, and even frequently assigned them to different species until we figured out that grasshoppers lay eggs that become locusts and vice versa.

    There’s some speculation about what the limits of non-genetic heritability are, but I think that the real point is that we assume different genes when we see different behavior and/or different appearance, but that’s not necessarily so. Plasticity can account for some of that, so why assume different genes before we get the tests back? Why not do more to investigate hypotheses related to how different environments lead to different gene expression? Why not look at things with different priorities and see what comes of it?

    No, I don’t think it’s a challenge to Darwin. I’m not even sure it’s a challenge to Dawkins. But it is a different way of looking at the problem: plasticity leads, genes follow is not how we would normally frame things.

    I don’t believe that this could possibly account for all evolution. There will be some times when novel mutation open up an organism’s new environment rather then helping exploit an organism’s existing (albeit recently changed or recently new) environment. In those cases, genes do lead. But if we don’t know to what extent environment can impact gene expression, and we don’t know to what extent a parent’s behavioral modeling (like the parent running to hunt for a child of a species previously living by ambush hunting) can impact gene expression, then there’s a fuzzy area of uncertain heritability and, should heritability exist, uncertain mechanism of heritability. He thinks this is a major direction research must take.

    What we find we won’t know until we take it. He plays up possibilities, but his examples of how we know things to operate are modest and, as you say, obvious. He tries to use those, however, to point to ways in which even the agreed upon facts as we now know them lead to conclusions about how evolution operates that are contrary to how researchers’ frame research questions.

    I’m not a researcher in the field. I don’t know if he’s right or wrong about this being contrary to research framing. But I think that’s what he’s saying, and he’s not wrong about the facts – plasticity does outpace genetic diffusion in the short run.

  82. 82
    PZ Myers

    Alf Hickey, #78:

    What separates the two fields is that change in biological evolution is causally driven by the natural selection, while there is no such process in geology.

    That’s like announcing that all of geology is driven by sedimentation, and that there is no such process in biology. Sure to make every knowledgeable person in the room stare at you for a bit.

  83. 83
    johnharshman

    gillt:

    I never intended to make an argument about non-genetic changes driving evolution.

    Well, if that’s so, you’re changing the subject. Better to announce such a thing in advance.

    I understand genotype to be a comparison of alleles (never mind coding or non-coding) between two or more individuals.

    It’s more like the sum of all alleles under discussion in some comparison. And if you think that “allele” doesn’t include isochores or chromosome number changes, it’s the wrong word, as a genotype can include any features of the genome you care to mention.

    Practically speaking, in phylogenetics this means making gene trees of SNPs but in the age of whole genome sequencing I see no reason not to accommodate isochores and other “higher order” genomic patterns, which you seem not to agree with, apparently.

    Certainly higher order patterns can be phylogenetic markers, but what does that have to do with what we’ve been talking about here? Nothing, as far as I can tell.

  84. 84
    johnharshman

    For the most basic example, consider the concentration gradients within the zygote that determine the first differential gene expressions in development. Where did that information come from? Nowhere in the embryo’s genome is that encoded. Was it encoded then in the mother’s genome.

    I see only two plausible answers. Either it was encoded in the mother’s genome (as I suspect is the case) or the gradient is a product of some physical feature of the egg, like yolk distribution. But in the latter case, either that causal feature is itself encoded in the genome, or it’s a universal feature of all eggs. That is, differences among species can’t be due to differences in distributional gradients in the egg unless those differences are passed down genetically. The gradients themselves are not inherited; they’re produced anew in each egg.

  85. 85
    gillt

    I think were more or less on the same page johnharshman, but I’ve never seen a chromosomes referred to as an allele before. Haplotype is probably a better word.

  86. 86
    johnharshman

    I didn’t use the word “allele”. You did. “Genotype” is my word, and that covers everything.

  87. 87
    gillt

    And if you think that “allele” doesn’t include isochores or chromosome number changes, it’s the wrong word, as a genotype can include any features of the genome you care to mention.

    I see. Is this what you meant to say? “….it’s [because] it’s the wrong word…”

  88. 88
    johnharshman

    Yes. Though I don’t see a difference.

  89. 89
    gillt

    Certainly higher order patterns can be phylogenetic markers, but what does that have to do with what we’ve been talking about here? Nothing, as far as I can tell.

    My contentions, whether you believe they have anything to do with what you’re talking about, are:

    1. Only recently have we been able to assess on a genome-wide scale these higher order patterns in the genome.
    2. Coupling a much wider array of phenotypes (DNA methylation patterns) and genotypes (methyltransferase genes) should provide more insight into evolutionary relationships, and better resolution of diversification and species histories. You can’t intuit the patterns from the sequence.
    3. This is a fundamental shift away from relying on just genes, as was mentioned in the OP, that will require new tools and new models.

  90. 90
    johnharshman

    I certainly agree with 1 and most of 2. But I think you can infer (not sure what “intuit” conveys) the pattern from the sequence. And 3 has nothing to do with the OP, which is not about phylogenetic analysis at all but about evolutionary innovation.

Comments have been disabled.