Yet another misguided attempt to revise evolution

What we have below (click on headline for free access) is a review in Nature by Denis Noble of a new book by Philip Ball, How Life Works: A User’s Guide to the New Biology, which has garnered good reviews and is currently #1 in rankings of books on developmental biology.  The Amazon summary promises that the book will revise our view of life:

A cutting-edge new vision of biology that will revise our concept of what life itself is, how to enhance it, and what possibilities it offers.

Biology is undergoing a quiet but profound transformation. Several aspects of the standard picture of how life works—the idea of the genome as a blueprint, of genes as instructions for building an organism, of proteins as precisely tailored molecular machines, of cells as entities with fixed identities, and more—have been exposed as incomplete, misleading, or wrong.. . .

I haven’t read it yet, though I will (I have several books ahead of it, including the galleys of Richard Dawkins’s new book, for which I’m to provide a blurb). Instead, I will review a review: Denis Noble’s review published a few days ago. (That’s the screenshot below.) Admittedly, it’s a review of a review, but Noble gives his take on the book’s importance, and in so doing reveals his own idea that neo-Darwinism is not only impoverished, but misguided in important ways.  And, as usual, Noble proves himself misguided.

In some ways it’s unfortunate that Noble was chosen as a reviewer, as the man, while having a sterling reputation in physiology and systems biology, is largely ignorant of neo-Darwinism, and yet has spent a lot of the last decade trying to claim that neo-Darwinism is grossly inadequate to explain the features and evolutionary changes of organisms. You can see all my critiques of Noble here, but I’ll just quote briefly from the latest to give you a flavor of how he attacks modern evolutionary theory:

In an earlier post I wrote, “Famous physiologist embarrasses himself by claiming that the modern theory of evolution is in tatters“, I emphasized five assertions Noble made in a 2013 paper in Experimental Physiology, and then I criticized them as being either deeply misguided or flat wrong. Noble’s claims:

1. Mutations are not random
2. Acquired characteristics can be inherited
3. The gene-centered view of evolution is wrong [This is connected with #2.]
4. Evolution is not a gradual gene-by-gene process but is macromutational.
5. Scientists have not been able to create new species in the lab or greenhouse, and we haven’t seen speciation occurring in nature.

I then assessed each claim in order:

Wrong, partly right but irrelevant, wrong, almost completely wrong, and totally wrong (speciation is my own area).

And yet Noble continues to bang on about “the broken paradigm of Neo-Darwinism,” which happens to be the subtitle of his new article (below) in IAI News, usually a respectable website run by the Institute of Art and Ideas.

And yes, Noble’s banging persists in his review of Ball’s book. The criticisms I level will be against Noble’s claims, as I can’t verify whether he’s accurately characterizing Ball’s views or spouting his (Noble’s) own misguided views.

The problem with Noble;s review is twofold: the stuff he says is new and revolutionary is either old and well known, or it’s new and unsubstantiated.  Here are a few of his quotes (indented and in italics) and my take (flush left):

First, Noble’s introduction to the book, which is okay until Noble tries to explicate it:

So long as we insist that cells are computers and genes are their code,” writes Ball, life might as well be “sprinkled with invisible magic”. But, reality “is far more interesting and wonderful”, as he explains in this must-read user’s guide for biologists and non-biologists alike.

On to Noble’s asseverations:

When the human genome was sequenced in 2001, many thought that it would prove to be an ‘instruction manual’ for life. But the genome turned out to be no blueprint. In fact, most genes don’t have a pre-set function that can be determined from their DNA sequence.

Well, the genome is more or less a blueprint for life, for it encodes for how an organism will develop when the products of its genome, during development, interact with the environment—both internal and external—to produce an organism.  Dawkins has emphasized, though, that the genome is better thought of as “recipe” or “program” for life, and his characterization is actually more accurate (you can “reverse engineer” a blueprint from a house and engineer a house from a blueprint—it works both ways—but you can’t reverse engineer a recipe from a cake or a DNA sequence from an organism.)  The DNA of a robin zygote in its egg will produce an organism that looks and behaves like a robin, while that of a starling will produce a starling.  You can’t change the environment to make one of them become the other. Yes, the external environment (food, temperature, and so on) can ultimately affect the traits of an organism, but it is the DNA itself, not the environment, that is the thing that changes via natural selection. It is the DNA itself that is passed on, and is potentially immortal. And the results of natural selection are coded in the genome. (Of course the “environment” of an organism can be internal, too, but much of the internal environment, including epigenetic changes that affect gene function are themselves coded by the DNA.)

As for genes not having a “pre-set function that can be determined from their DNA sequence,” this is either wrong or old hat.  First, it is true that at this point we don’t always know how a gene functions from its DNA sequence alone, much less how it could change the organism if it mutates. This is a matter of ignorance that will eventually be solved. As for “pre-set function”, what does Noble mean by “pre-set”?  A single gene can participate in many developmental pathways, and if it mutates, it can change development in unpredictable ways, and in ways you couldn’t even predict from what that gene “normally” does. The gene causing Huntington’s chorea, a fatal neurodegenerative disease, has a function that’s largely unknown but is thought to affect neuron transport. But it also has repeated sections of the DNA (CAGCAGCAG. . . . .), and mutations that increase the number CAG repeats can cause the disease when they exceed a certain threshold.

But the “Huntington’s gene” is not there to cause disease, of course. It interacts with dozens or even hundreds of other genes in ways we don’t understand. What is its “pre-set” function? The question is meaningless. And was does “pre-set” mean, anyway?

The second sentence in the bit above is garbled and ambiguous, and at any rate doesn’t refute the notion that the genome is indeed the “instruction manual for life.”

But wait: there’s more!

Instead, genes’ activity — whether they are expressed or not, for instance, or the length of protein that they encode — depends on myriad external factors, from the diet to the environment in which the organism develops. And each trait can be influenced by many genes. For example, mutations in almost 300 genes have been identified as indicating a risk that a person will develop schizophrenia.

It’s therefore a huge oversimplification, notes Ball, to say that genes cause this trait or that disease. The reality is that organisms are extremely robust, and a particular function can often be performed even when key genes are removed. For instance, although the HCN4 gene encodes a protein that acts as the heart’s primary pacemaker, the heart retains its rhythm even if the gene is mutated.

“Polygeny,” or the view that traits can be affected by many genes, is something I learned in first-year genetics in 1968. But some “traits” or diseases are the product of single genes, like the trait of getting Huntington’s Chorea of sickle-cell disease.  But many diseases, like high blood pressure and heart disease, can be caused by many genes. And it’s not just diseases. Whether your earlobes are attached to your face or are free is based on a single gene, and eye color, to a large extent, is too (see this list for other single-gene alternative traits).

As far as the HCN4 gene goes, mutations may allow it to have a rhythm, but many mutations in that gene cause abnormal rhythms.and can even bring on death through heart attacks. No, the gene is not robust to mutations, and I can’t understand where Noble’s statement comes from. It appears to be wrong. (I am not attributing it to Ball here.)

More:

Classic views of evolution should also be questioned. Evolution is often regarded as “a slow affair of letting random mutations change one amino acid for another and seeing what effect it produces”. But in fact, proteins are typically made up of several sections called modules — reshuffling, duplicating and tinkering with these modules is a common way to produce a useful new protein.

This is not a revision of the “classic” view of evolution because we’ve known about domain-swapping for some time. For example, the “antifreeze” proteins of Arctic and Antarctic fish can involve changes in the number of repeats in the enzyme trypsinogen, which normally has nothing to do with preventing freezing. Or, antifreeze proteins can arise via the cobbling together of bits of different known genes, or from bits of the unknown genes, or even be transferred via horizontal acquisition from other species.  Yes, this happens, but it’s not the only way by a long shot that evolution occurs. In fact, now that we can sequence DNA, we’ve found that many adaptive changes in organisms are based in changes in single genes or their regulatory regions, and not swapping of modules. Here’s a figure from a short and nice summary by Sarah Tishkoff from 2015 showing single genes involved in various adaptations that have occurred in one species—our own. The traits are given at the top, and the genes involved are by the symbols. For example, though several genes can involve skin pigmentation, mutations in just one of them can make a detectable change.

At any rate, we can nevertheless regard shuffling of domains (or even horizontal gene transfer from other species) as mutations, and the new mutated gene then evolves according to its effect on the replication of the gene. No revision of neo-Darwinism or its mathematics is involved. New ways of changing genes haven’t really revised our view of how evolution works, even when we’re talking about the “neutral theory” instead of natural selection.

These mutations, by the way, contra Noble, are still “random”—that is, they occur irrespective of whether they’d be useful in the new environment—and although they can make big changes in the organism’s physiology or appearance, can nevertheless evolve slowly.  A gene with a big effect need not evolve quickly, for the rate of evolution depends not on the effect on the organism’s appearance, physiology, and so on, but on its effect on the organisms’s reproductive capacity. And these things need not be correlated.

Later in the book, Ball grapples with the philosophical question of what makes an organism alive. Agency — the ability of an organism to bring about change to itself or its environment to achieve a goal — is the author’s central focus. Such agency, he argues, is attributable to whole organisms, not just to their genomes. Genes, proteins and processes such as evolution don’t have goals, but a person certainly does. So, too, do plants and bacteria, on more-simple levels — a bacterium might avoid some stimuli and be drawn to others, for instance. Dethroning the genome in this way contests the current standard thinking about biology, and I think that such a challenge is sorely needed.

Ball is not alone in calling for a drastic rethink of how scientists discuss biology. There has been a flurry of publications in this vein in the past year, written by me and others^2–4. All outline reasons to redefine what genes do. All highlight the physiological processes by which organisms control their genomes. And all argue that agency and purpose are definitive characteristics of life that have been overlooked in conventional, gene-centric views of biology.

This passage verges on the teleological.  For surely organisms don’t have “goals” when they evolve.  If a mutation arises that increases the rate of replication of a gene form (say one increasing tolerance to low oxygen in humans living in the Himalaya), it will sweep through the population via natural selection. If it reduces oxygen binding, it will be kicked out of the population. Can we say that increased oxygen usage is a “goal”? No, it’s simply what happens, and I suspect there are other ways to adapt to high altitude, like getting darker skin. To characterize organisms as evolving to meet goals, as Noble implies here, is a gross misunderstanding of the process.

Yes, the organism is the “interactor”, as Dawkins puts it: the object whose interaction with its environment determines what gene mutations will be useful. But without the “replicator”—the genes in the genome—evolution cannot occur.  The whole process of adaptation, involving the interaction of a “random” process (mutation) and a “deterministic” one (natural selection), is what produces the appearance of purpose. But that doesn’t mean, at least in any sense with which we use the word, that “purpose” is what makes organisms alive.

But the appearance of “purpose” as a result of natural selection brings up another point, one that Dawkins makes—or so I remember.  I believe that he once defined life as “those entities that evolve by natural selection.”  I can’t be sure of that, but it’s as good a definition of life as any, as it involves organisms having replicators, interacting “bodies”, and differential reproduction. (According to that definition, by the way, viruses are alive.)  So if you connect natural selection with purpose, one might say, “Life consists of those organisms who have evolved to look as if as if they had a purpose.”  But I prefer Dawkins’s definition because it’s more fundamental.

At the end, Noble says that this “new view of life” will help us cure diseases more readily:

This burst of activity represents a frustrated thought that “it is time to become impatient with the old view”, as Ball says. Genetics alone cannot help us to understand and treat many of the diseases that cause the biggest health-care burdens, such as schizophrenia, cardiovascular diseases and cancer. These conditions are physiological at their core, the author points out — despite having genetic components, they are nonetheless caused by cellular processes going awry. Those holistic processes are what we must understand, if we are to find cures.

I haven’t heard anybody say that “genetics alone can help us treat complex diseases”. You don’t treat heart disease by looking for genes (though you can with some cancers.) But genetics can surely help! For genetic engineering is on the way, and at least some diseases, like sickle-cell anemia, will soon be “curable” by detecting the mutated genes in embryos or eggs and then fixing the mutation with CRISPR. And advancesin genetics are surely helping us cure cancer—see this article.  But of course some diseases, even those with a genetic component, need environmental interventions: so called “holistic” cures. There may, for example, be a genetically-based propensity to get strep throat. But if you get it, you don’t worry about genes, you take some penicillin or other antibiotic. (Curiously, the form of Streptococcus that causes strep throat doesn’t seem to have evolved resistance to the drug!)

Overall, I don’t see much new in Noble’s take on evolution—just a bunch of puffery and regurgitation of what we already know. Perhaps people need to know about this stuff in a popular book, but, after all, Noble’s piece was written for scientists, for it appears in Nature.

Despite repeated claims in the last few years that neo-Darwinism is moribund or even dead, it still refuses to lie down. Happy Darwin Day!

Addendum by Greg Mayer: For those interested in the distinction between the blueprint (wrong) and recipe (on the right track) analogies for the genome, I wrote a post explicating the difference, citing and quoting Richard, here at WEIT; the post also explains why the Wikipedia article about “Epigenetics” is definitionally wrong; see especially the link to this paper by David Haig.