Introducing Geremy (and Behe)

Nothing you have posted is relevant to anything I’ve asked about. Nobody is saying that mechanical signals don’t affect gene expression. But these mechanical signals are not inherited by gametes or zygotes. They can’t explain the differences among species.

The biggest misconception that I am pushing up against here is that evolutionary theory has ever had a real explanation for phenotypic differences as opposed to anecdotes, analogies and just so stories. This narrative that the theory is pushing that genotypic evolution is somehow creating new phenotypes, and the constraints of ontogeny are irrelevant is both very old and demonstrably false. What is more it is a result of how evolutionary theory was originally designed, here is how one source that is very friendly to the current theory explains it:

The founders of the modern synthesis treated embryology as a ‘black box’, the details of which could be ignored for the purposes of evolutionary theory; their focus was on the transmission of genes across generations, not the process by which genes make organisms. This strategy was perfectly reasonable, given how little was understood about development at the time.

Population Genetics (Stanford Encyclopedia of Philosophy) Section 4

So what about evolutionary theory today, does it actually address the feasibility of evolving actual organisms, complete with the known soft matter physics based constraints on ontogeny? Or does it still only create complicated speculations about how genes can be transmitted from taxa to taxa, while ignoring the very real constraints on such processes? The answer seems to be that modern evolutionary theory tries to incorporate developmental biology but is simply unable to for reasons that I will now explain.

Any theory of common descent will be constrained by the mechanisms of biological descent. So there is no rational basis to believe that evolutionary theory can ignore the constraints of most developmental processes or simply cherry pick a few elements of ontogeny such as the role of HOX genes, while ignoring how the overall process works and still explain how such development processes can change to create new phenotypes. Unlike what occurs on the organismal level where predators and environmental changes can be accurately described by the metaphor of natural selection, the constraints on developmental processes are imposed by physiology and soft matter physics. Physiological adaptations can dramatically change the phenotype of animals as a result of plasticity, but on the genetic level it most constrains genetic variation (I will talk more about this point later). Soft matter physics simply is a property of the universe, as is all physics and does not change in response to the needs of organisms. So when I describe processes by which emergent mechanical forces regulate both gene expression and phenotype shape, I am describing the role of soft matter physics in development as demonstrated by empirical evidence from developmental biology, mechanobiology and tissue engineering. Unlike genetics where information is contained in nucleotide sequences, the constraints imposed by physics under a given circumstance, do not need to be encoded anywhere in particular because they exist everywhere in the entire visible universe, so simply recreating the exact same circumstances recreates the same physical constraints.

Understanding how mechanical pressures caused by cell movements create predictable differences in the shape of the cell, their nuclei and chromosomes arrangement in different cell types helps us to understand both the role of non coding DNA in making such arrangements and the importance of genome topography in gene regulation. So what I have been explaining is that it is the topography of the entire genome that varies slightly even between individuals of the same species, and varies much more between different species that is causing most the differences in phenotype. What is more it is these differences in whole genome shape and positioning that causes the distribution of mechanical forces within cells to change, resulting in differential transcription in different organs of the same organism, different members of the same species and different species. This physic based understanding of gene regulation is the result of advances in technology, not theoretical considerations. Here is how one study explains it:

Most studies examining three-dimensional structure genome-wide are limited to a single species. In this study, we compared three-dimensional structure in the genomes of induced pluripotent stem cells from humans and chimpanzees. We collected gene expression data from the same samples, which allowed us to assess the contribution of three-dimensional chromatin conformation to gene regulatory evolution in primates. Our results demonstrate that gene expression differences between the species may often be mediated by differences in three-dimensional genomic interactions. Our data also suggest that large-scale chromatin structures (i.e. topologically associating domains, TADs) are not well conserved in their placement across species. We hope the analytical paradigms we present here could serve as a basis for future comparative studies of three-dimensional genome organization, elucidating the putative functional regulatory loci driving speciation.

So there is plenty of evidence that is consistent with the hypothesis that mechanical pressure that arises during development regulating gene expression. This hypothesis is made stronger when we look carefully at the alternative hypothesis that gene controlled genetic regulatory networks are globally regulating gene expression.

Some problems with the current model of gene regulatory networks were delineated in a paper entitled “Rethinking gene regulatory networks in light of alternative splicing, intrinsically disordered protein domains, and post-translational modifications”, the author explains:

Models for genetic regulation and cell fate specification characteristically assume that gene regulatory networks (GRNs) are essentially deterministic and exhibit multiple stable states specifying alternative, but pre-figured cell fates. Mounting evidence shows, however, that most eukaryotic precursor RNAs undergo alternative splicing (AS) and that the majority of transcription factors contain intrinsically disordered protein (IDP) domains whose functionalities are context dependent as well as subject to post-translational modification (PTM). Consequently, many transcription factors do not have fixed cis-acting regulatory targets, and developmental determination by GRNs alone is untenable. Modeling these phenomena requires a multi-scale approach to explain how GRNs operationally interact with the intra- and intercellular environments.

If we combine both the evidence that the genetic elements of GRNs are characterized by indeterminacy, with the observations that originally inspired the GRN hypothesis, namely the that there was a bi-stable genetic switch that activated the lytic cycle bacteriophages that had previously been in the lysogenic phase, it seems obvious at least to me, that the stochastic regulatory elements of the genome are being organized, by a highly structured outside force not the intrinsic genetic elements. The mechanical pressures that arise from cell movements are a natural consequence of the soft matter physics that universally constrains intracellular, and intercellular organization of all multicelluar organisms on earth.

The evidence that gene regulation is a result of physics, has implications for evolutionary theory:

  1. It removes natural selection as a flexible factor of gene regulation evolution. Natural selection is a metaphor for environmental constraints that change with time. There is no evidence that soft matter physics has changed, since the initial appearance of water in the universe about 11 billion years ago.
  2. As some biologists knew decades ago, repetitive gene elements are not junk DNA, but rather create the sequence dependent structural context for gene expression. The well know rule of, form equals function applies to DNA, just as much as it applies to protein.

Word salad too. Whether sequence similarities are caused by common descent has nothing to do with cell movements but with the structure of the comparative data, a consistent nested hierarchy. We know what causes changes in DNA sequences: various sorts of mutations whose mechanisms are fairly well understood. You have to answer this simple question: why a nested hierarchy?

The nested hierarchies of evolutionary theory are strictly conceptual models used to make biological complexity comprehensible to most people who are linear thinkers. What I am saying is that the actual variables that the nested hierarchy model is used to explain, do not change in the manner predicted by neutral theory they fail at their principle objective of demonstrating that the evidence is consistent with common decent as opposed to common physiological constraints. A good example of this is rRNA sequences used in some phylogenetic trees, there is good evidence going back several decades that rRNA differences between eukaryotic species is not explainable by the neutral theory while that of bacteria and archaea superficially appears to be. That simple observation was made in the paper below:

If we go look at the evidence for why rRNA in prokaryotes appears to evolve in line with the neutral theory, it gradually becomes clear that the appearance of neutral evolution is deceptive. The real answer seems to be that the variation rRNA gene sequences are determined by shared physiological constraints. Here is how one paper explains it:

We discovered that rRNA alleles are differentially expressed in response to nutrient limitation-induced stress and that the rRNA allele most upregulated on a relative basis is distinguished by conserved sequence variants clustered in the small subunit head domain of the assembled ribosome (Kurylo et al. 2018). These findings revealed for the first time that the rRNA allele composition of the actively translating ribosome pool is indeed regulated in response to physiological stimuli. Remarkably, we further showed that ribosomes bearing these sequence variations causally affected stress-response gene expression and phenotype, including biofilm formation, cell motility, and antibiotic sensitivity. Consistent with the sequence variants modulating the so-called ribo-interactome (Simsek et al. 2017), we further identified rRNA allele-dependent genetic interactions with stress-related proteins that transiently interact with the ribosome in the proximity of the sequence variants during protein synthesis. The varied expression of this operon alters the mechanisms of ribosome-interacting factor engagement during translation to affect gene expression and cell physiology during stress. Interestingly, the exact same sequence variations were found to be conserved in many Enterobacteriaceae, including Salmonella enterica, which diverged from E. coli more than 120 million years ago.

Now if we compare the above example to humans we can say that it there is strong evidence all of us carry many rRNA s that have coding sequences that diverge from the standard human rRNA sequences used in phylogenetic trees. For example human chromosome 21 has 101 different rRNA coding sequence variants as explained in the paper below:

So I think that there are probably rRNA gene variants in one animal species that are identical to the dominant rRNA gene sequence of another species, but that the dominant rRNA gene variant of given species is dependent on the constraints that the physiology of a given species places on rRNA gene sequence variation . So what this would mean for phylogenetic trees that use rRNAs to represent genetic distance, is that the researchers are likely to be simply comparing how similar the physiological constraints on ribosomes gene sequences of selected species are to each other, not their genetic distance from each other due to common descent.

Its been awhile since I considered this, but I do not believe that complete rRNA sequences are used for phylogenetic studies. Rather, only relatively small regions are used. It may be that these small regions may include sequences that have allelic variants as suggested by @Geremy, but probably not.

Just something to keep in mind.


Just not true. Are you at all familiar with evolutionary developmental genetics (“evo-devo”)? Further, it’s clear that genetic differences result in phenotypic differences, even though we’re not always sure how. And you are still faced with the problem that these physical factors in development are not inherited.

That quote doesn’t mean what you’re trying to make it mean.

Don’t know, because I have no idea what that pile of words means.

What are the very real constraints on gene transmission, by which I assume you mean inheritance between generations?

Careful. This is edging toward the common creationist trope of “science doesn’t know everything, therefore science knows nothing.”

But aren’t all organisms under the same constraints? How, then, can those constraints explain the differences among species?

I don’t think that’s true, but even if it is, isn’t the topography a result of the sequence, so changes in topography are caused by changes in sequence?

I don’t believe that’s true either. Evidence suggests that most alternative splicing is noise, non-functional splicing errors.

But why are cells moving in that way, and differently in different species?

No, it doesn’t. It just introduces an intermediary between DNA sequence and development. Clearly, phenotypes are subject to selection. You can’t deny that. This is true regardless of how genotypes influence phenotypes.

A very few such elements are known to have functions. Most of them clearly do not. Are you familiar with the onion test?

That’s an excuse that allows you to ignore all the comparative data. Good luck with that. What follows is more word salad. You never seem to tire of it.

You understand that phylogenetic analyses are not just measuring genetic distance, right? Why should physiological constraints follow a nested hierarchy? Why do trees made from other sequences match those from rRNAs so well? What are the differences in physiological constraints among crocodylians that affect exons, introns, and 3’ UTRs in exactly the same way so as to produce the same nested hierarchy?


There’s no misconception going on here. All you’re doing is making assertions, and then wasting a lot of space with overly wordy arguments that don’t lead to the conclusion you are selling.

You also completely neglected to actually answer @John_Harshman’s point, which is that:

But these mechanical signals are not inherited by gametes or zygotes. They can’t explain the differences among species.

You just ignored this and went on to write hundreds of irrelevant words, and give references that does nothing to rebut or answer the point.

Who the hell claims that “the constraints of ontogeny are irrelevant”? That’s right, nobody claims that. Of course, it is a demonstrable reality that genetic changes result in novel phenotypes. You are literally ignoring a century of research in genetics here. The new phenotypic effects of millions of mere genetic mutations are known.

I have referenced articles in this very thread that shows how genetic mutations have resulted in new phenotypes in dog breeds, affecting attributes from cranial morphology to the size of their legs. You completely ignored that post.

The phenotypic influence of genetic mutations on everything from the chemistry catalyzed by enzymes, to the patterns of camouflage on insects (and many other animals), the direction and patterns of cell division in multicellular tissues, to the colors of bioluminescent animals, have all been directly demonstrated by experiment.

Genetic changes don’t result in new phenotypes? That is a claim so fatuous it should be beneath even bothering with a response. Seriously. That is so far on the outside of fringe nutbaggery you might as well be arguing the Earth is flat.

How is anyone to take you seriously now?


Nobody has ever claimed that all the difference that accumulate in genetic sequences between species (be it in rRNA or other genes) are neutral. Your argument here literally does not make sense.


Well, most of them are, and that’s certainly a very common claim. Of course most sequences are neither rRNA nor genes, and large parts of most eukaryote genes are introns.

Infertility. Lack of money for a wedding ring. Birth Control. Distance between the male and female. It could be lots of things.

1 Like

Genome topography is part of the story, yes. As you said yourself, this varies with species and as such is evolvable, leading to topographical divergence in distantly related species that influences gene regulation and sometimes leads to phenotypic differences. It’s genetic changes, in the same way that direct point mutations in a regulatory element are.

You’re saying that the shape of the genome in the nucleus causes mechanical forces that somehow lead to differential tissue-specific expression? How? Sounds like you’re missing a step. All cells in a body have the same genome present in their nuclei, so there would have to be some initial change that led to different genome shapes and positions in different tissues, and only then would you get differential tissue-specific gene expression.

You quoted from this study
And then said:

but that study has nothing at all to do with mechanical pressures regulating gene expression. You might also want to re-read a few sentences from the first paragraph of the introduction:

The molecular mechanisms that underlie regulatory adaptation have also been the focus of much research. Studies in mice, flies, yeast, and primates have revealed that expression divergence between species is often driven by mutations or epigenetic modifications within cis-regulatory elements (CREs)

The existence of these kinds of studies is a problem for your idea that the sequence of regulatory elements is all but irrelevant to gene regulation.

Nothing about the paper you cited here disagrees that GRNs control development, they just argue that this process is not 100% deterministic and that development is of course also influenced by some context-specific factors.

This is your problem - mechanical forces in development are not “highly structured” enough to be the primary determinants of development. The 2015 paper by Niklas et al you cited earlier was criticising the narrow view of GRNs being purely deterministic, but now you’ve just gone to the other extreme by saying that GRNs are actually not deterministic at all, it’s actually mechanical forces that are controlling everything.


I know that some on this board have suggested that I should not cite videos, however I’m going to cite a physics video to explain the science of mechanical pressures, by using the concept of tensors. Tensors are used by physicists and engineers to predict the effects of the laws of physics on three dimensional space. I first learned how biology worked on the molecular level by using tensors to predict the existence of molecular structures and then researching the biochemical and genetic basis of those structures. Only afterward did I realize that there are free molecular biology textbooks online, but I’m glad that I studied molecular biology the way that I have (as an autodidact I am still a student) using tensors because the text books that I have read don’t mention tensors. So first I will let someone who is able to explain the mathematics of tensors much better than I can, explain what a tensor is:

Next I will allow a couple of developmental biologist explain how tensors can be used to explain the three dimensional structure of a cell:

Cell tensional forces that are generated in the actomyosin cytoskeleton are also resisted by these adhesive tethers to neighboring cells and the ECM, and by internal cytoskeletal structures that resist being compressed, such as microtubules or cross-linked actin filaments within filopodia (Wang et al., 2001; Brangwynne et al., 2006; Ingber, 2006). This results in the establishment of a tensegrity-based mechanical force balance that generates cytoskeletal prestress ensuring cell, tissue and organ shape stability (reviewed by Ingber, 2006).

Because cells and tissues are tensionally prestressed, forces transmitted between cells across ECM and cell-cell junctions are transmitted to multiple sites in the cell over cytoskeletal linkages, and the efficiency of this signal transfer is sensitive to the level of

tensional prestress in the cells, which ‘tunes’ the cellular response, much as mechanical tension tunes a violin string (Box 1) (Wang et al., 2001; Wang et al., 2009). Mechanical forces exerted on surface adhesion receptors and channeled along cytoskeletal filaments can be converted into changes in molecular biochemistry at the cell surface or in junctional complexes (e.g. focal adhesions, cell-cell adhesion complexes) via tension- or strain-dependent changes of molecular shape that expose new binding sites, as seen, for example, in p130Cas, talin and fibronectin (Tamada et al., 2004; Sawada et al., 2006; Brown and Discher, 2009; Zhong et al., 1998; Gee et al., 2008; del Rio et al., 2009), that alter molecular binding or unbinding kinetics (Lele et al., 2006b) or that modulate ion flux through membrane channels (Thodeti et al., 2009). These forces might also be concentrated at distant sites in the nucleus, where they can alter gene activity or stem cell fate through the modulation of nuclear ion channels (Itano et al., 2003) or through the nuclear transport of key signaling molecules, such as of b-catenin (Farge, 2003; Kahn et al., 2009), or of transcription factors, such as TFII-I or Gata2 (Mammoto et al., 2009), to determine cell fate and drive tissue development. Again, the efficiency of force transmission depends directly on the level of isometric tension in the cytoskeleton (reviewed by Ingber, 2006; Wang et al., 2009). Importantly, such a hard-wired mechanism for nuclear mechanotransduction in response to forces acting on the cell surface is much faster than conventional chemical signal propagation (Na et al., 2008).

I would highly recommend that everyone who is dubious of the role in biology that I am attributing to mechanical forces read all of the papers that I cite in full, and perhaps study a little soft matter physics when they get a chance.

Now the reason that mechanical pressures have such a predictable effect on the cytoskeletal structures is that they all involve the cooperation of two very stable proteins, that cooperate not demonstrably due to any shared evolutionary relationship, actin and tubulin. Here is what one molecular biology textbook has to say about this relationship:

It is remarkable that actin and tubulin have both evolved nucleoside triphosphate hydrolysis for the same basic reason—to enable them to depolymerize readily after they have polymerized. Actin and tubulin are completely unrelated in amino acid sequence: actin is distantly related in structure to the glycolytic enzyme hexokinase, whereas tubulin is distantly related to a large family of GTPases that includes the heterotrimeric G proteins and monomeric GTPases such as Ras, structures that are discussed in Chapter 3.

The importance of the cytoskeleton is hard to overstate. Here is how the same developmental biology paper that I mentioned earlier explained it:

The myriad findings described above illustrate the point that mechanical forces play a central role in morphogenesis and tissue patterning, and that physical cues are as important as chemical factors for developmental control throughout virtually all stages of embryogenesis. Although the contribution of physical forces to cell and tissue deformation in the embryo has been recognized for over a century, it is only recently that mechanical stresses have been shown to function as informative signals that produce specific changes in molecular biochemistry and gene expression through the process of cellular mechanotransduction. Moreover, although virtually all physical forces in the developing embryo must, at some level, result from the action of its constituent cells, it is only now becoming clear that cytoskeletal tension is the driving force behind many of these key mechanical processes and mechano-chemical transduction events. Cells sense changes in mechanical signals based on their ability to alter biochemistry and to induce remodeling in the cytoskeleton and the ECM at the molecular level, but the extent of this response and the efficiency of mechanical signal transfer throughout the cell and nucleus depend on the level of isometric tension or prestress in the cytoskeleton. At the same time, cell- generated tensile forces alter the chemical signals that are conveyed by cells, in addition to producing the distortion of neighboring cells and ECM molecules that propagate mechanochemical signaling over long distances, thus driving tissue patterning and organ formation at the scale of the whole embryo.

So the evidence best supports the hypothesis that it is deformations in the cytoskeleton cause by mechanotranduction, that globally regulates gene expression in all multicellular species on earth. If you would like a more up to date explanation of how I would suggest reading the mechanobiology paper linked below:

So as I stated earlier gene network regulation in all multicellular organisms it is a result of soft matter physics. Theories that attempt to claim that non teoleogical changes to the genome can globally change gene regulation, have to posit realistic mechanisms capable of changing the shape of the entire genome, without appealing to changing selection pressures because physics based constraints are not as flexible as the imaginations of population geneticists, and require engineering of some sort to work around them. As one Evo Devo paper stated years ago:

First, traditional comparative approaches to the evolution of development—whether focused on the morphological or on the molecular/genetic level—are reaching their limits in terms of explanatory power. The more we learn about the evolution of pattern-forming gene networks, or the ontogeny of complex morphological traits, the more it becomes clear that it is less than straightforward to conclude anything about evolutionary origins or dynamics based on such comparisons alone…

The second deadlock concerns the integration of ecology or, more precisely, the active role of the environment in phenotypic evolution. Over the last few decades, it has become increasingly clear that genes and genetic programs are simply not sufficient to explain the ontogeny of most morphological traits (see, for example, Goodwin 1982; Oster and Alberch 1982; Nijhout 1990; Alberch 1991; Webster and Goodwin 1996; Keller 2000; Pigliucci 2010). Instead, a more interactive view has emerged—treating genes and their organismic as well as external environment as influencing each other in a regulative feedback loop (e.g., Waddington 1957; West-Eberhard 1998, 2003; Odling-Smee et al. 2003; Kirschner and Gerhart 2005; Gerhart and Kirschner 2007; Gilbert and Epel 2009; Moczek 2012). In this view, the environment is not just passively endured by an organism, determining its chances of survival. It plays an active and essential role in development through phenotypic plasticity (West-Eberhard 2003; Gilbert and Epel 2009), and is itself altered by the activity of the organism (Odling-Smee et al. 2003). An obvious example of the latter is humanity’s ability to massively change and manipulate the environment to our own (short-term) liking and comfort. A number of useful concepts, such as facilitated evolution (Kirschner and Gerhart 2005; Gerhart and Kirschner 2007), genetic accommodation (West-Eberhard 1998, 2003, 2005a, b), and niche construction (Odling-Smee 1995; Laland et al. 1999; Odling-Smee et al. 2003) have been proposed to tackle this challenge, but a unifying and rigorous framework to deal with the active role of the environment in developmental evolution is still missing (Moczek 2012). (Boldness added for emphasis)

I am in agreement with the writers statements above, but I don’t share their optimism that an undiscovered evolutionary mechanism exists that can make evolution happen as imagined by evolutionary biologists. So you can imagine what I think when someone asserts that such mechanisms are known to exist already. I also think that mechanobiology is only going to further reduce the explanatory power of evolutionary theory by explaining the constraints of evolution in the language of physics and mathematics rooted in real world observations as opposed to the speculative conceptual models of evolutionary theory.

Once again one must ask how this physics is inherited so as to create the developmental differences among species.


Here’s a simple challenge for you, Geremy: Explain how genotypes and alleles have nothing to do with the results of Mendel’s pea experiment.



Mechanotransduction can influence gene expression, but that’s very different from “controlling” gene expression. As I said before, the mechanical forces are just nowhere near precise or specific enough to do what you want them to do. Take the example cited in the 2010 review you linked by Mammato and Ingber - Itano et al., 2003.

They showed that mechanical influences on the nucleus caused by cell spreading can increase the level of calcium ions in the nucleus. This can indeed influence gene expression, as they showed, by activating calcium-dependent enzymes that go on to activate certain transcription factors. This is inevitably going to influence a lot of transcription factors, so it lacks the kind of fine-scale control required in developmental gene expression networks. That fine control can only be encoded in regulatory sequences, which can then change and evolve.

You keep ignoring the elephant in the room, the evidence that the very papers you cite bring up for sequences changes being key for many morphological changes, in some cases precisely because they have a downstream effect on the mechanical properties of cells. No one is disagreeing that mechanical forces don’t have an effect on gene expression or development, so continually citing papers to support that obvious point isn’t helping you make your larger argument.


Hey @Geremy it would be really nice if you could proceed to start answering questions some time soon.

I mean, instead of just mindlessly referencing papers that don’t support your claim that genetic mutations don’t explain changes in organismal phenotypes.

It’s the same thing every time. You make a claim about how mechanical forces affect gene expression (which nobody here disagrees with), then reference papers that say so. Okay, good.

But then you proceed to make grandiose declarations about the inability of genetic mutations to affect heritable phenotypes which simply do not follow from the papers you reference.

And you keep ignoring questions that go straight to the heart of your conclusions, and you ignore references that straightforwardly contradict the declarations about evolution that you make.

Are you here to just copy-paste large volumes of technical literature for the sake of appearance(look at all this material you’re quoting that must look really clever and technical to someone who haven’t got a clue), or are you interested in actually discussing the merits of your grandiose declarations?


We don’t need any undiscovered mechanisms; I do predict that new, minor ones will be discovered. The ones we have account for the vast majority of evolution.

Geremy, please try to step back for a moment. If someone who has studied the cytoskeleton, particularly the actomyosin cytoskeleton, for decades thinks that your claims are overblown and based on a lack of depth in your approach, have you considered the possibility that you may be wrong?


[quote=“Geremy, post:125, topic:13373”]
What I think I said was that if one is explaining the origins of the phenotype then genetics has a limited explanatory power, or something to that effect. Perhaps I should explain what I meant. As you no doubt know some traits are simple traits where a simple genetic mutation could reshape it easily, while many complex traits are spread over much of the genome. So it’s not so simple to claim that any particular genes are the cause of any particular phenotype instead they are associated with them.

PubMed Central (PMC)

An expanded view of complex traits: from polygenic to omnigenic

I have no doubt that gene evolution can impact the phenotype to a small degree by turning the expression of a given gene on or off, etc. What I don’t think is that such a process can reconfigure the topography of the whole genome, which sparked my interest in gene regulatory networks.

A couple of years ago I read an old article by James Shapiro about the inability of natural selection to explain most of evolution, and after investigating his claims about it I decided that his argument against selection seemed well grounded. Here is his old article:

His critique is what made me switch my pet project of understanding developmental biology better, to one of understanding the intersection between developmental biology and evolution.

It’s just a rebranding. Please note that his rhetoric hasn’t motivated Shapiro or anyone else to actually do anything.

The whole genome doesn’t have to be reconfigured to explain the vast majority of phenotypic changes. TAD boundaries are generally quite conserved, most of the changes between animal groups, for example, involve smaller scale intra-TAD rearrangements, which can easily be generated by recombination and/or sequence changes leading to altered protein binding sites. This isn’t some kind of big mystery.

Please show that “reconfiguration of the topography of the whole genome”(whatever that is even supposed to mean) is something that is required in the history of life.

1 Like

Short Answer

The reason higher order 3 D regulatory elements positioning is not conserved, is because their positioning is genetic sequence dependent. Specifically it is the sequences of the so-called junk DNA sequences which are functionally divergent between species is what determines the relative stiffness and flexibility of the chromatin in chromosomes controlling the positioning of the chromosome coils and supercoils in the nucleus. This precise positioning allows for both differential gene regulation in a cell type specific manner, which is determined by mechanical pressure regulated nucleus shape. It also causes very divergent gene regulation between at least humans and chimpanzees that have very similar genome coding sequences but divergent non coding sequences.

Long answer


The authors of the study that I mentioned earlier were of the same opinion that you were before they did their study, so they reviewed the previous literature and found that the evidence for TAD are highly conserved in their placement is weak. They wrote:

Thus, based on our analysis of the literature, we believe that the common notion that TADs are highly conserved in their placement across species is not well supported. Indeed, recent evidence from yeast [89], different Drosophila tissues [90], and plant species [52] suggests that TADs and TAD-like domains may not be particularly conserved, which raises questions about the stability of these higher-order structures and the significance of their role in the evolution of gene regulation across different lineages.

Some of the problems with the earlier literature that they found was that previous studies hadn’t actually performed an assessment of the whole genome but inferred TAD conservation based on the functional conservation of TAD boundaries in specific organs. Where as the studies that had compared the whole genome had compared species where they had more data for one species than they had for the other one, by comparing the proportion of TADs that were conserved in the the species for which they had more data that could also be found in the species for which they had less data. The author thought that methodology inflated the similarities between species which seems to be a reasonable conclusion to me.

Now to fully understand how important this is, we have to understand the the shape of chromosomes in the nucleus between divisions is not random, but is determined by the sequences of the so called junk DNA, which as everyone points out is not conserved between species. It is these long repetitive sequences that cause some parts of the chromatin to be more stiff than others, which allows the DNA to coils to form in non random patterns in stem cells. Here is how one paper describes ths:

In summary, our results indicate that sequence composition is a key aspect of chromatin TADs and hub formations. Other large-scale correlates, such as gene density and protein-DNA binding affinities, also contribute to spatial organization and local concentrations around nuclear bodies. The initiation step for TAD or loop formation is under “compositional constraints”, essentially driven by local flexibility or stiffness of the coiled DNA fibre. In such a context, intrinsic properties of DNA sequence, bendability, and binding affinity of promoters and enhancers, may have a strong influence on TAD dynamics and the phase separation behaviour of chromatin. The formation of active chromatin assemblies is compositionally biased and may take place in both GC-rich and GC-poor chromosomal environments, but gains strength in mechanically soft regions (GC-rich), where DNA-protein foci coalesce via multivalent links. Interactions among and within chromatin domains can be viewed as part of a flexible “chromatin code”79 that can help in deciphering to what extent the non-coding space of contemporary genomes is “junk”80,81 or “polite”82.

Notice that the writers are very aware that their research might invalidate the concept of junk DNA, so I am not reading into the data that which is not there.

Moving on,the specific organization of the genome in the nucleus allows for changes in the replication rate of cells in a cell type specific way as explained in another paper which states:

Eukaryotic chromosomes replicate in a temporal order known as the replication-timing program1. During mammalian development, at least half the genome changes replication timing, primarily in units of 400–800 kb (“replication domains”; RDs), whose positions are preserved in different cell types, conserved between species, and appear to confine long-range effects of chromosome rearrangements27. Early and late replication correlate strongly with open and closed chromatin compartments identified by high-resolution chromosome conformation capture (Hi-C), and, to a lesser extent, lamina-associated domains (LADs)4,5,8,9. Recent Hi-C mapping has unveiled a substructure of topologically-associating domains (TADs) that are largely conserved in their positions between cell types and are similar in size to RDs8,10. However, TADs can be further sub-stratified into smaller domains, challenging the significance of structures at any particular scale11,12. Moreover, attempts to reconcile TADs and LADs to replication-timing data have not revealed a common, underlying domain structure8,9,13. Here, we localize boundaries of RDs to the early-replicating border of replication-timing transitions and map their positions in 18 human and 13 mouse cell types. We demonstrate that, collectively, RD boundaries share a near one-to-one correlation with TAD boundaries, whereas within a cell type, adjacent TADs that replicate at similar times obscure RD boundaries, largely accounting for the previously reported lack of alignment. Moreover, cell-type specific replication timing of TADs partitions the genome into two large-scale sub-nuclear compartments revealing that replication-timing transitions are indistinguishable from late-replicating regions in chromatin composition and lamina association and accounting for the reduced correlation of replication timing to LADs and heterochromatin. Our results reconcile cell type specific sub-nuclear compartmentalization with developmentally stable chromosome domains and offer a unified model for large-scale chromosome structure and function.

So all of the above supports the hypothesis that coding genes and non-coding genetic sequences are both equally important to biological function, so in order to change how coding genes are regulated one needs a hypothesis of how to either gradually or rapidly transform the sequences of the entire genome. Now I will place the above information in its proper developmental context.

Soft matter physics and the phenotype

I have just a few point to make in section:
My first point is that the positioning of such regulatory elements before the gastrulation is random due to the nucleus behaving like a liquid, as supported by the paper below:

Secondly, by using mouse models scientists have learned that the shape of the blastocyst is shaped into a sphere using the physics of bubble formation as you can read here:

My next point is that by using both fruit flies and zebrafish embryos scientists have demonstrated that it is the cell movements in gastrulation that causes the chemical pathway/ gene transcription cascade that is usually used to explain gastrulation. The paper states:

We started by comparing zebrafish and drosophila, that belong to the two main branches of the bilaterian evolutionary tree (Deuterostomia and Protostomia). In both species, we used assays that first blocked gastrulation movements, and then performed exogenous deformations to artificially rescue mechanical strains. Interestingly, in both species, we found that blocking gastrulation movements blocked the expression of certain developmental genes, and that rescuing movements reestablished it. Moreover, both the identity of the responding cells and the identity of the responding genes turned out to be similar in both species: in both cases, the responsive cells were the presumptive mesoderm, and mechanical signals activated key early transcription factors for early mesoderm specification – notail in zebrafish (a brachyury orthologue) and twist in drosophila. Finally, in both species, the mechanotransduction pathway turned out to be the same: cell deformation in the presumptive mesoderm promotes phosphorylation of β–catenin at a conserved site (tyrosine 654 in zebrafish and 667 in drosophila), which in turns promotes its translocation into the nucleus, where it acts as a transcription factor and turns on mesoderm genes.

Now I will use one last paper to explain that it is the fluid-like quality of stem cells, that allows the positioning of the chromosomes to be adjusted by the compressive forces of gastrulation, in by the reshaping of the nucleus, a process that is repeated during tissue migration and organogenesis. The part about gastrulation is described in the 2017 mechanobiology paper that I mentioned in my earlier post. In Box 1 it states in part:

A number of recent studies have correlated the positions of various post-translational modifications, transcription factor binding sites, enhancer–promoter loops and RNA polymerase II occupancy with Hi-C data in order to gain insights into the spatial dimension of gene regulation106–108. Chromosome organization models have been introduced to address the mechanical coupling between nuclear morphology and gene expression146. In these models, the spatial arrangement of chromosomes is viewed as a configuration of ellipsoids (the chromosome territories) packed into an ellipsoid-shaped container (the nucleus). The shape of the container is defined by mechanical constraints, and cell type-specific configurations are determined by solving an optimization problem, where the pairwise overlap between two chromosomes is penalized on the basis of their difference in gene expression levels. The solutions to this optimization problem are configurations that link nuclear morphology with chromosome organization and gene expression97,146.

So there is plenty of evidence, way too much for me to explain in this post that supports the central role of mechanical pressure in the global regulation of genetic networks. Mechanical pressure during development molds the shape of the nucleus and its compartments in a cell specific way, which allows the chromosomes which folds in a nucleotide sequence specific way to fold in such a way that some parts of the genome are accessible to transcription while other parts are not. Rather than overturning genetics I think that by highlighting how important the gene sequence of the entire genome is it elevate genetic to one of a few crucially important tools of proper cell function.

Incidentally: onion test.