The Unz Review - Mobile

The Unz Review: An Alternative Media Selection

A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
 Gene Expression Blog
Epigenetics Does Not a Revolution Make

Email This Page to Someone

 Remember My Information


51x-WwY-sAL._SX387_BO1,204,203,200_ Periodically I get frankly stupid comments that seem to imply that the incredible swell of results coming out of molecuar genetics and genomics are revolutionizing our understanding of evolutionary and population genetics. Over the past generation it’s been alternative splicing, then gene regulation and evo-devo, and now epigenetics is all the rage. The results are interesting, fascinating, and warrant deeper inquiry (I happen to see graduate school admission applications for genetics, and I can tell you that conservatively one out of three applicants mention an interest in epigenetics; the hype is grounded in reality, as epigenetics may be a pretty big deal in human health that we can effect).

But these fields don’t tear down the bigger picture of evolutionary and population genetics in terms of what they teach us. It is not a revolution (see Evolution in Four Dimensions: Genetic, Epigenetic, Behavioral, and Symbolic Variation in the History of Life for a dissenting take, but note that the authors are minority voices). Unfortunately the belief that everything has changed is widely held. I can understand why and how social scientists who wish to downplay classical heritability of traits would latch upon epigenetics (it’s this decades’s gene-environment interaction), but even biologists outside of genetics have told me in conversation that they assumed that the fields of evolutionary and population genetics has been “revolutionized” and the textbooks had to be “rewritten” (these are real words/phrases). Now mostly it’s because of epigenetics.

My patience with this sort of thing is minimal at this point. I’ve had to deal with it so much that I’ve written several “why genetics has not been revolutionized” posts. And here’s another. The media certainly isn’t helping by hyping so much. Rather soon we’ll see someone writing about how epigenetics is “disrupting” evolutionary biology like Uber is disrupting transportation! About ten years ago I remember trying hard to convince a producer on a public radio show out of doing a segment about how epigenetics was changing everything we knew about evolution. I just didn’t believe it was the case. He was frustrated, but thankfully he didn’t just go and just find people who wanted to say what he would have preferred to have been said on air. (and I’m sure he checked in with several people and they all came back with the same feedback)

The most recent edition of the closest thing that I know of to a Bible of population genetics, the venerable Hartl and Clark text, is nearly ten years old. In the last edition there was a chapter on “population genomics,” a field which really didn’t exist for the earlier editions. But I doubt that “population epigenomics” will be added for the next edition. Not because it isn’t potentially something important or research-worthy, I just don’t think that it will be a necessary part of every population geneticist’s toolkit (in contrast, pretty much all population geneticists who work with empirical data are going to become genomicists, if they aren’t already).

41PHSZN6AEL In science the word “revolutionary” is a big deal. Genomics has not revolutionized evolutionary and population genetics, so I definitely don’t think epigenetics has revolutionized these fields. There’s a reason that The Genetical Theory of Natural Selection, written by R. A. Fisher in the 1920s, is still useful reading for someone interested in evolutionary and population genetics. Fisher’s fusion of Mendelian genetics with evolutionary theory (along with the efforts of Sewall Wright and J.B.S. Haldane) allowed for the development of a formal field which could extend beyond verbal logic and empirical description. Perhaps the second major revolution to identify in the 20th century would be the emergence of molecular methods in assaying genetic variation after the work of Lewontin and Hubby. Though genomics has resulted in a quantitative increase of data on the orders of magnitude beyond what was available with allozymes in the 1960s, my own judgment is that the top-level inferences from the earliest results using coarse molecular markers have not been overturned as much as refined and specified. Genomics allows for a scaling up of the empirical possibilities, but it is a matter of extension more than doing something new under the sun.

5190MX7FYQL._SY344_BO1,204,203,200_ As with many things much of the confusion has to do with semantics. In his book The Darwin Wars Andrew Brown makes a distinction between thinkers who conceive of an ‘analytic gene’, as opposed to the more concrete sort favored by molecular geneticists today. By the latter, I mean a specific sequence mechanistically bound together in what we would today term a ‘genic region,’ transcribed and (a subset in many cases) translated into proteins. Though those with a fidelity toward the latter definitions often imply that their views are more concrete and adhere more to reality, it is important to note that the foundations of Mendelian thought are fundamentally about analysis as opposed to mechanism. There is a reason that a core textbook for introductory genetics is titled Genetic Analysis. Genetics began as inferences about the nature and character of inheritance from observed patterns, not by understanding molecular biological mechanisms. Mendelian genetics flourished 50 years before the final understanding of its molecular basis in DNA. Evolutionary biology emerged as a field 50 years before genetics as we understand it emerged.

Why? Let me quote two passages. First, from Dan Dennett in Darwin’s Dangerous Idea:


Darwin’s ideas about the powers of natural selection can be lifted out of their home base in biology. Indeed, as we have already noted, Darwin himself had few inklings (and what inklings he had turned out to be wrong) about how the microscopic processes of genetic inheritance were accomplished. Not knowing any of the details about the physical substrate, he could nevertheless discern what if certain conditions were somehow met, certain effects would be wrought. This substrate neutrality has been crucial in permitting the basic Darwinian insights to float like a cork on the waves of subsequent research and controversy, for what has happened since Darwin has a curious flip-flop on it. Darwin, as we noted in the preceding chapter, never hit upon the utterly necessary idea of a gene, but along came Mendel’s concept to provide just the right structure for making mathematical sense out of heredity (and solving Darwin’s nasty problem of blending inheritance). And then, when DNA was identified as the actual physical vehicle of the genes, it looked at first (and still looks to many participants) as if Mendel’s genes could be simply identified as particular hunks of DNA. But then complexities began to emerge; the more scientists have learned about the actual molecular biology of DNA and its role in reproduction, the clearer it becomes that the Mendelian story is at best a vast oversimplification. Some would go so far as to say that we have recently learned that there really aren’t any Mendelian genes. Having climbed up Mendel’s ladder, we must now throw it away. But of course no one wants to throw away such a valuable tool, sill proving itself daily in hundreds of scientific and medical contexts. The solution is to bump Mendel up a level and declare that he, like Darwin, captured an abstract truth about inheritance. We may, if we like, talk of virtual genes, considering them to have the reality distributed around in concrete materials of the DNA)….

51k+5+6j9OL._SY344_BO1,204,203,200_The Origin of Species is a rich and worthwhile read even today 150 years after its publication. Evolutionary ideas are as old as Western philosophy, and they were in the air during Darwin’s time. The reason we remember his theory is that it had a precise rigor and mechanism attached to its explanations of the empirical reality around us. In particular, the concept of natural selection driving adaptation upon heritable variation. A major lacunae, as noted above, was that Charles Darwin did not posit any plausible mechanism of maintaining variation. The attempts in The Origin of Species were reaches, and from what I recall there were multiple shifts in emphasis across the editions of this book. Without discrete particulate Mendelian inheritance the variation that was the raw material for natural selection disappeared. But observe that all that was necesssary was a system of inheritance where variation was maintained. If on an alien planet the substrate of inheritance was different in fundamental molecular configuration from DNA one would still be able to posit a theory of evolution in whose general outlines are Darwinian, because the ultimate input of heritable phenotypic variation would remain the same.

Second, W. D. Hamilton in Narrow Roads of Gene Land: Volume 1:

…I had made the decision that I would not even try to come abreast of the important work that was being done around me on the molecular side of genetics. This might well be marvelous in itself: I admitted the DNA story to concern life’s most fundamental executive code. But, to me, this wasn’t the same as reading life’s real plan. I was convinced that none of the DNA stuff was going to help me understand the puzzles raised by my reading of Fisher and Haldane or to fill in the gaps they had left. Their Mendelian approach had certainly not be outdated by any of the new findings.

51XHYubQzNL._SY344_BO1,204,203,200_ In case you are not aware of who W. D. Hamilton was, he was arguably he most influential evolutionary biologist of the second half of the 20th century (there are other contenders, but Hamilton’s name has to be in the mix). The idea of inclusive fitness was the fruit of his inquiry into the “problem of altruism” (a problem that the famed medical geneticist Lionel Penrose summarily dismissed as worthy of financial support according to Hamilton, who had a project relating to chromosome biology lined up for his potential mentee). As it happens molecular genetics, or more precisely its descendant, genomics, has taken some interest in Hamilton’s ideas even if he didn’t take quite an interest in it. The ubiquity of selfish genetic elements may be understood as an extension of Hamiltonian inclusive fitness dynamics at the intra-genomic level (there are other arguments, see Michael Lynch’s Origins of Genome Architecture). And many of the predictions that Hamilton’s formalism made in regards to the nature of the origins of sociality and sex are best explored with molecular genetic assays.

51zeajUmWhL._SX316_BO1,204,203,200_ Finally, I want to bring to your attention R. A. Fisher’s 1941 paper Average Excess and Average Effect of a Gene Substitution (the link is not gated). Fisher was writing before DNA. His conception of a gene was analytic by necessity (though by his period it was understood that genes were resident on chromosomes). That is, he was imagining a unit of inheritance characterized by alleles. Today we often think of a defined genetic sequence as this unit of inheritance, and alleles are usually assumed to be changes on a single base pair of DNA, a single nucleotide polymorphism (though there are other types of genetic variants, such as copy number variants). But these are not necessary to work out the basics of evolutionary genetics, as is clear from the fact that Fisher, Haldane and Wright managed to do so before comprehension of mechanistic details (though as a physiological geneticist Wright thought more mechanistically, and that might explain why he was right and Fisher was wrong in regards to the reason for the existence of genetic dominance). Fisher used terms like “allelomorphs”, and many of the characters he was familiar with would have been tracked through correlations of phenotypes. In an abstract and fundamental sense an allele is just a variant segregating in the population. It could be a SNP, or a CNV, a microsatellite, or an indel. Or it might also be a regulatory element. Fisher and many of his colleagues only postulated the allele after seeing the association between phenotypic marker and the novel trait (known markers had positions on the genetic map); they were often ignorant of the detailed biophysical basis for the variation.

41ZhyEU5lGL._SY344_BO1,204,203,200_Average effects and average excesses, keys to quantitative genetics, and underlying Fisher’s model of evolutionary change over time, have within them the richness to absorb the myriad mechanistic details cascading out of modern molecular genetics and genomics. If you read him closely it is clear that Fisher did not assume the sort of deterministic relationship that some of his critics impute to him. He understood penetrance, as did all geneticists long before the vagaries of gene expression and epistatic interaction were elucidated in their mechanistic details. But over the long haul the average effect of a substitution in the population is critical. Understanding the nature of the average effect gets you much of the way in this game, if not all the way.

There are some Christians who assert that their religion is the natural completion of Judaism and Greek philosophy.* There are others who rather argue that Christianity was a radical revolution against all that came before. Historically the latter has been a minority view. The Marcionites failed, and the Jewish origins of Christianity were sewn into the fabric of its foundational scripture in the form of the Old Testament. And despite periodic revolts, the reality is that intellectual Christianity speaks with a Greek philosophical voice. Ultimately this debate is of purely academic interest for me. But it exhibits a similarity with academic arguments and debates. In Endless Forms Most Beautiful: The New Science of Evo Devo Sean B. Carroll takes a traditionalist approach which suggests that novel 61DFNJkqyGL._SX331_BO1,204,203,200_ results from the new field of evolutionary developmental biology firmly supports and extends the Neo-Darwinian Synthesis. Carroll’s book is under 400 pages. It is elegantly written and economical of prose, and it proposes an evolution in our thinking about the nature of the variation which serves as the raw material for natural selection. Contrast that with the late Stephen Jay Gould’s The Structure of Evolutionary Theory, which came in at nearly 1,500 pages. Published in the early 2000s, much of it was written earlier. There are only two references to epigenetics within it. If Gould had not died in 2002 he would probably have come out with a new revised edition by now, and I’m rather confident that epigenetics would loom very large indeed. Though Sean B. Carroll is a very eminent scientist, he remains a bit player on the public intellectual scene. That’s because he does not promise revolution, he comes bearing a twist on the orthodoxy. In contrast, Gould’s prolix prose was rich with the promise of paradigms shattered and lost, and grand visions of heretics risen up to prophetic status, as the statues of the grand old men of the Neo-Darwinian orthodoxy were torn down to make way for the new idols (this old Paul Krugman slap at Gould is pretty on point about why he was so popular in the 1990s). Reality is more prosaic than intellectual revolts plotted in used bookstores!

Carroll’s ended with a quotation from Charles Darwin because his espousal of a particular theory in regards to evolution was in no way contradictory to the spirit of The Origin of Species. Endless Forms Most Beautiful was a paean to Darwinism, properly conceived. Charles Darwin was no dogmatist in regards to the origin of variation, though he was blind to the possibility that Mendel’s experiments provided. I doubt he would have taken much umbrage at Sean B. Carroll’s update to the canon, as its genetic nature postdated him by decades. Some of the original Mendelians had arrayed themselves against the biometrical school, which considered itself a custodian of Charles Darwin’s thought during the period when classical evolutionary theory was somewhat in decline (see The Eclipse of Darwinism: Anti-Darwinian Evolution Theories in the Decades around 1900). But as told in Will Provine’s The Origins of Theoretical Population Genetics the conflict was short lived, and the synthesis which emerged from the debates of that period laid the basis for modern evolutionary biology, a field far richer and more robust than during Darwin’s own time.

And I may be wrong here as I’m no historian of Watson and Crick’s discovery, but I don’t see that they thought of themselves as overthrowing Mendelianism, as opposed to putting it on a firmer molecular and biophysical basis. A comprehension of the biological machinery within the synaptonemal complex only enriches and extends our understanding of the nature of Mendelian process.

Quantum mechanics was a revolution in physics because it introduced a whole domain of understanding which operates outside of the purview of classical physics, parallel to it to this day. The modern project of unification and reconciliation continues and is unfinished. In contrast Darwinian evolution, Mendelian genetics, and molecular genetics extend and complement each other. If quasi-Lamarckian heritable epigenetic patterns within the genome were so powerful and ubiquitous as to overturn a Mendelian understanding of heritability, then the Mendelian model of inheritance would not have been so persuasive and crystal clear in the first place in the analyses of the Fly Room. If our understanding of evolution and genetics was contingent on perfect understanding of the molecular mechanisms and machinery by which evolutionary processes occur, then we’d have been at a loss before 1952 (in actually, 1952 was only the start in any case). We weren’t in the wilderness, because understanding can manifest itself at multiple layers of abstraction and complexity. Just as a deeper understanding of neuroscience can only benefit psychology, so a deeper understanding of biophysical phenomena such as epigenetics will only enrich our understanding of evolutionary and population level dynamics. There is no revolution in evolution. At least until we get better with CRISPR….

* This is a general trend. Some Chinese Christians have argued that the religion completes and complements Confucianism, while Karen Christians point to similarities between Christianity and indigenous religious beliefs.

• Category: Science • Tags: Epigenetics, Genetics

8 Comments to "Epigenetics Does Not a Revolution Make"

Commenters to Ignore
Endorsed Only
[Filtered by Reply Thread]
  1. The great irony here is that epigenetics is being absorbed and incorporated into evolution and population genetics, not upturning it.

    So much of the epigenome is shaped by the underlying sequence that my first assumption when looking at epigenomic differences is that there is a causative genetic variant. Proving that something is a “pure” epiallele as defined by Eric Richards is incredibly difficult and few studies actually do the necessary work to show that an epiallele is independent of genetic variation.

    Working in epigenetics, it is incredibly frustrating how loosely people use the term to mean even things like differential gene expression or to be simultaneous with the environment. It has led to a rash of bad studies that fail to adequately control for genetic variation (I could name names) and grave misunderstandings not only amongst the public, but increasingly scientists too. Don’t even get me started on Eva Jablonka and her crowd. They operate in some sort of dream land.

    Reply More... This Commenter Display All Comments
  2. The gene was first understood as a discrete inheritable unit and under that definition epigenetics is still about genes. It was only when the gene was narrowly defined as a locus of DNA when we ended up with both genetics and epigenetics.

    • Replies:
    Reply More... This Commenter This Thread Hide Thread Display All Comments
  3. Under Waddington’s original definition ~1942 it pertained to the idea of how cells differentiated and became locked into that state, passing it on to new cells while having the genetic material. He saw it as a regulatory mechanism. Arguably this is concurrent or even before it was really conclusive that DNA was the hereditary substance. So…I find this comment a bit confusing.

    It should also be noted, that technicaly, prions represent “epigenetic” inheritance. Something I think it is under-appreciated. While that is a rather exceptional case, its an exception that really helps put a definition of epigenetics into context.

    This perspective published in Genetics: “What Do You Mean, ‘Epigenetic’” should be required reading for the field. Does a great job of cutting through all the confusing and misleading definitions of the term:

    • Replies:
    Reply More... This Commenter This Thread Hide Thread Display All Comments
  4. Somewhat related: I kinda feel like the gut microbe buzz is the new epigenetics type hyperbole but I hope I’m wrong.

    Reply More... This Commenter Display All Comments
  5. Indeed, the meaning of “epigenetic” changed as well since the 1930′s.

    Reply More... This Commenter This Thread Hide Thread Display All Comments
  6. Evo devo in the real world – Pharyngula
    • Website     Show CommentNext New Comment

    […] I disagree with Razib Khan on a lot of things, but he’s exactly right on recent fads in biology. […]

    Reply More... This Commenter Display All Comments
  7. Evo devo in the real world
    • Website     Show CommentNext New Comment

    […] I disagree with Razib Khan on a lot of things, but he’s exactly right on recent fads in biology. […]

    Reply More... This Commenter Display All Comments
  8. Anonymous
    • Website     Show CommentNext New Comment

    Although I would agree that epigenetics does not make an evolutionary revolution, the author goes too far. Once one understands what the neo-Darwinian synthesis actually meant, it is clear that it was undermined by results molecular evolution.


    Unfortunately, the change is obscured by goal-post-shifting. To know what the neo-Darwinian synthesis actually stood for, you have to be ancient like Masatoshi Nei, or you have to study the issue, as I have [1]. The rest of you, alas, are victims of an unrelenting campaign of distortion that shifts the meaning of “neo-Darwinism” to include what Darwin and his historic followers clearly rejected, while presenting only straw-man alternatives, reinforcing the meme that There Is No Alternative.

    In reality, the conceptual synthesis of Mendelian inheritance, mutation and selection was developed by Bateson, Morgan, Punnett, Johannsen and so on, mostly while little Ronny Fisher was a schoolboy. Fisher, et al later rejected their view, because– although it recognized a crucial role of selection in adaptation– it invoked mutation as a source of initiative, creativity, direction, and discontinuity. These are non-Darwinian ideas, and by that I mean that these ideas are explicitly treated as alternatives and often labeled “anti-Darwinian” or “non-Darwinian” in historic sources. For instance, Morgan allowed that sometimes evolutionary change involved the acceptance of a single mutation with a distinctive effect, a saltation. If evolution could happen by single mutations, then tendencies of mutation could result in trends or parallelisms.

    The architects of the Modern Synthesis were committed Darwinians, so they conjured up a more restricted version of Mendelian evolutionary genetics that acted more Darwinian, and insisted that this must be the way things actually work. They insisted that evolution was always a multi-locus process of smoothly shifting the frequencies of alleles generated by recombination, so that selection never had to wait for a new mutation. They insisted that change had to be gradual, and that variation could only be a source of abundant “raw materials”. They insisted that the rate of evolution would not show a close relationship to the mutation rate. [2] This has all gone down the memory hole.

    Biologists today accept non-Darwinian ideas such as saltations and mutational trends (e.g., GC content). What has changed is simply that they put the label “Darwinian” on anything that sounds reasonable. Orr refers to the mutational landscape model as a model of “Darwinian adaptation”. Instead it is a model of de Vriesian aptation by acceptance of single mutations. The architects of the Modern Synthesis did not even call this kind of process “adaptation”– they called it “random pre-adaptation” [3].

    To put this as briefly as possible, you do not understand the Modern Synthesis if you think it resulted from a bottom-up process of teasing out the evolutionary implications of genetics, selection and mutation. Instead, it resulted from a top-down process of trying to find a Mendelian way to rationalize pre-existing Darwinian commitments. The resulting speculative “gene pool” theory of evolutionary genetics was declared to be universal, but began to fall apart almost immediately due to results from molecular evolution.


    Reply More... This Commenter Display All Comments

Comments are closed.