|
ix This book is about the origins of novelty in evolution.
p.1 Introduction, A Clock on the Heath
p.8 we possess 22,500 genes, only six times the number possessed by a bacterial cell, the simplest of all known free-living organisms. How could human complexity be achieved with so few genes? ...The answer... lies in the multiple use of versatile conserved components... The living organism... is complex in unusual ways appropriate for versatility and modification rather than for dedicated single use.
[JLJ - A more recent source cites 20,500 genes in the human genome. Other sources count genes in other ways and end up with twice this value.]
p.9 Mutation only changes what already exists. It does not create new anatomy, physiology, and behavior from nothing, so we need to know how readily one structure can be transmuted into another, particularly when we consider structures of intricate design and interdependent activities.
p.10 Chapter 1: The Sources of Variation
p.13 There are limits on what selection can accomplish. We must remember that it merely acts as a sieve, preserving some variants and rejecting others; it does not create variation.
p.14 Given the remote connection between the DNA and the phenotype, we have no way of knowing how often random DNA modification can produce useful outcomes for selection.
p.31 According to Gould, Darwin thought that variation must meet "three crucial requirements: copious in extent, small in range of departure from the mean, and isotropic" (or undirected toward adaptive needs of the organism). Gould called these three attributes of variation Darwin's most brilliant insight, "because he realized that selection could not otherwise operate as the creative force in the evolution of novelties."
p.34-35 all organisms are a mixture of conserved and nonconserved processes (said otherwise, of unchanging and changing processes)... Novelty in the organism's physiology, anatomy, or behavior arises mostly by the use of conserved processes in new combinations, at different times, and in different places and amounts, rather than by the invention of new processes... Central to our argument is that these processes, many of which have been conserved for hundreds of millions or even billions of years, have very special characteristics that facilitate evolutionary change. They have been conserved, we suggest, not merely because change in them would be lethal... but because they have repeatedly facilitated changes of certain kinds around them.
p.35 We will talk later about the concept of weak regulatory linkage, which means essentially that links between processes can be forged without extensive retooling of each component. To maintain these links, processes are often reinforced with additional weak linkages - the suspender and belt approach to reliability.
p.36 We argue that regulatory components are the main targets for heritable change... Sewall Wright, the great population geneticist, said it most clearly: "The old writers on evolution were often staggered by the seemingly necessity of accounting for the evolution of fine details..., for example, the fine structure of all the bones... Structure is never inherited as such, but merely types of adaptive cell behavior which lead to particular types of structure under particular conditions." It is remarkable that in 1931 Wright could foresee a time when it would be possible to explain anatomy and physiology in terms of the cell's adaptive responses to differing conditions. We will show that such adaptability is built into most of the cell's conserved core processes.
p.37 The road to evolutionary change is paved with physiological adaptability. Phenotypic variation, and along with it evolutionary change, is facilitated by simple regulatory tweaks to existing physiological and developmental processes that long ago were designed so that the organism could adapt to its environment.
[JLJ - If you already have a process that makes a cake, and you want a "new" kind of cake, all you should have to do is randomly vary one of the steps in making the cake, you then just let the cake-making process bootstrap itself to completion. When the 'once-modified cake making process' is done 'making the cake', you ought now to have a variation on a type of cake.]
p.38 Chapter 2: Conserved Cells, Divergent Organisms
p.39 Novelty usually comes about by the deployment of existing cell behaviors in new combinations and to new extents, rather than in their drastic modification or the invention of completely new ones.
p.70 After the fixation of core processes, evolution seems to have proceeded at a steady if not increasing pace. The conserved processes are called core processes because they are deeply involved in generating the phenotype... Mere inability to change does not explain their long-term persistence, because other processes could in principle arise, surpass, and replace them. Why do they persist for such long evolutionary times? ...If they are under continuous selection, what are they being selected for?
p.71 Chapter 3: Physiological Adaptability and Evolution
p.72 what is selected is not simply a specific state of a biological system, but more commonly mechanisms that can produce a range of states in response to a range of conditions.
p.106 The lesson from Baldwin, Schmalhausen, and Waddington is that the organism has a great deal of latent novelty within its own somatic adaptability. As West-Eberhard has extended the lesson, all phenotypic novelties are reorganizations of preexisting phenotypes. In effect, the organism can express many alternative phenotypes - phenotypes that are stable... or phenotypes that are readily reversible... Because they have already been tested in evolution, these phenotypes are necessarily viable and adaptive to the ambient conditions.
p.107 Adaptability is a key characteristic of many of the conserved core processes of eukaryotes.
p.107 The organism... Its robustness stems from a physiology that is adaptive.
p.108 The secret of the stability of the phenotype is dynamic restoration... Evolution can achieve new forms of somatic adaptation so readily because the system, at all levels, is built to vary.
p.109 Chapter 4: Weak Regulatory Linkage
p.109-110 These two facts, conservation and economy, suggest that complexity must arise through the multiple use of a relatively few conserved elements. Complexity arises when different parts of the adaptive range are selected. It also arises when different combinations of conserved elements are chosen.
p.111 In this chapter we use the term weak linkage... to mean an indirect, undemanding, low-information kind of regulatory connection, one that can be easily broken or redirected for other purposes.
p.136 The two meanings of the word "weak" in weak linkage, referring to reconfigurability and to unstable interaction, underlie the permissive and switch-like behavior of many biological processes.
p.143 Chapter 5: Exploratory Behavior
p.143-144 A subset of these core processes is what we call exploratory processes, or core processes that display exploratory behavior. Their adaptability is central to their function.
p.144 Exploratory behaviors... generate many, if not an unlimited number of, specific states in the course of their function, and provide a mechanism for selecting among these states those that best meet the particular pysiological need. Because they produce so many states, the cell will not use most of them; but under new selective conditions these can generate novel structures... Although exploratory processes make use of weak linkage, they use it to choose a small number of alternatives from a large number of possibilities.
p.145 trial and error plays a major role in the behavior of an entire organism. For example, when we look at how ants forage for food and learn to exploit their discoveries, we find a trial-and-error strategy that bears an eerie resemblance to the means by which the cytoskeleton self-organizes.
p.145 Some of the answers to both the present and future complexity of multicellular organization are found in exploratory processes based on randomness and functional selection. The nervous system thereby can construct itself with a relatively small number of rules. The plasticity afforded by physiological variation and selection not only accounts for much of how the organism generates the complexity of the nervous system or other forms of anatomy, but also for how these systems repair damage... It also helps explain how new anatomies can evolve from existing forms. Exploratory behavior is especially evident in conserved processes operating in the spatial dimension.
The processes for generating physiological variation and selection are themselves complicated. In some cases, the organism goes to great lengths to generate variation at each phase of a physiological adaptation. This capability is seen most clearly in the vertebrate adaptive immune system, which is based on exploratory principles.
p.146 The vascular system uses exploratory mechanisms to respond to local needs. It generates limited variation and achieves its final structure by selective stabilization. Such a system can grow with the individual, can vary to meet demand, and can easily change during evolution.
p.146 It is the evolutionary role of exploratory processes that causes us to give them such prominence in this book. They seem to be able to overcome barriers to novelty, since they generate novel structures in the course of their normal physiological function. This competency addresses the problem of evolutionary adaptations that require simultaneous events.
p.146 Might processes that generate significant variation in their routine function also reduce drastically the number of steps to achieve novelty?
[JLJ - A random number generator creates endless variation, but little novelty... novelty is simply produced as the output of a novelty-producing process... of whatever form or complexity required. Novelty requires a seed, an idea, a sketch, which sets up a chain reaction, continuing until stopped, or a novella is produced...]
p.147 it is now believed that bats are most closely related to whales, dogs, and deer.
[JLJ - A mammal in its core processes is 'poised to modify' to adapt to the environment. Clearly, the common ancestor of the bat and the whale had the potential to adapt into either creature - the 'variation potential' present within the internal structure and conserved processes must have been immense... Perhaps this idea can be applied to game theory - we seek positions with a huge 'potential to adapt' to whatever the environment looks like down the road, in the uncertain future. We seek simply for our variation and adaptation potential to be greater than that of our opponent, then we play a move our heuristics suggest - and then wait for the environment to reveal what was hidden, to sculpt our adaptive potential into something real...]
p.148 The architecture of cells is achieved without an architect... Cell shape responds to developmental and environmental cues independently of genetic control.
[JLJ - A trick that works simply works, whether it was by design, whether it can be backed by theoretical proofs, or whether it is simply being executed blindly by a structure that is designed to simply execute.]
p.149 Part of the process of achieving cell organization relies on trial and error, a form of physiological variation and selection at the level of protein assembly.
p.149 The key to the adaptability of microtubles is their dynamics.
p.150 microtubules... continually grow, disintegrate, and regrow, each microtubule persisting for only five minutes... Microtubules extend randomly from their tips and depolymerize back to the nucleation center by loss from their tips... They continue doing so until they encounter a stabilizing activity in the cell periphery, which blocks depolymerization at the tip... Microtubules that randomly enter the region of stabilization persist
p.152 Cytoskeleton formation is an exploratory process; many potential cell shapes are generated from a single genotype, even under stable environmental conditions. The cell can adapt to any signal that stabilizes any of its numerous potential phenotypes. The mechanism of microtubule turnover does not terminate the resulting arrangement of microtubles. Instead, cell organization is driven by stabilizing agents acting peripherally in the broadly responsive and unbiased process of microtubule assembly.
p.155 From these rigid rules emerges a highly adaptive strategy applicable to changing environments.
[JLJ - If you have a highly adaptive exploratory process 'attached' to something coltrolled by a mutation, the mutation gets modified by such a process - to the point that the overall effect could eventually become 'smoothed out' and 'refined' for usefulness.]
Microtubule assembly and ant foraging are conceptually analogous. Both are exploratory processes involving variation and somatic selection.
p.155 In the case of microtubules, the ends of certain microtubules are prevented from shrinking back. Other microtubules turn over, while a few each time are recruited and stabilized; as a result, the entire array becomes redirected to a new configuration.
p.155 Exploratory variation and selection together are powerful tools for generating physiologies and behaviors that are not merely extensions of existing behaviors. In both processes what the genome encodes is the means to explore, not the outcome of the exploration.
p.156 Evolution is about life and death, Malthusian growth and survival.
p.157 In the nervous system, cell death prunes away superfluous nerve cells after an exloratory process has made tentative connection with potential targets.
p.157 Neurons... These cells extend their long, thin axons somewhat randomly into the periphery of the body, like foraging ants. If an axon tip by chance enters the anatomically appropriate region, it receives survival factor produced there by target tissues, and it persists. If it enters the wrong region, it receives no survival factor and commits suicide.
[JLJ - Who says a "search" has to be followed by a specific "selection"? We see here that an exploration can be followed by a general survival criteria, with the remainder of the set "perishing".]
p.171 exploratory processes lower the hurdle for generating novelty
p.173 Muscle precursor cells are formed in the trunk in clusters close to the nerve cord. From this site, they migrate outward and follow an exploratory path into the neighboring appendage. There they associate with the bones and cartilage in whatever arrangement they find. The muscle precursors then proliferate and differentiate in response to local cues. Hence, migration of skeletal muscle precursors into the developing limb is an exploratory process, much like microtubule assembly.
p.173 As described earlier, the nerve axons also follow an exploratory path into the developing limbs... Finally, the vascular system sends migrating cells out to furnish vessels to any region that does not get enough oxygen.
p.174 the normal development of limbs holds the key to the rapid evolution of new limbs. Normal development begins with the patterned deposition of cartilage-forming cells, the precursors of the bones. Then follows a series of highly adaptive processes that can generate the muscle-nerve-blood-vessel anatomy of the normal limb, but can also develop any of an unlimited number of related states, defined by the location of the bones. Thus, initially only the skeletal elements of a limb may respond to genetic change and the other tissues can adapt to them... Exploratory processes... have an immense breadth of adaptation. Their adaptability is used in each organism in its normal development and in wound healing and regeneration. Thus, these broadly adaptive processes are under continuous selection for the function they serve and are available to support evolutionary change, when needed.
p.176 Exploratory mechanisms have a dual role in facilitating evolutionary change... By being globally responsive and adaptive they blunt the effects of mutation and reduce its effect and lethality. In this way they make possible the persistence of novel changes by reducing collateral damage, thus increasing the amount of heritable variation.... On the other hand, exploratory systems can be the targets of both environmental and genetic change.
p.177 Chapter 6: Invisible Anatomy
p.178 When a trait of anatomy changes in evolution, it is really the development of that trait that has changed. Anatomy itself is not inherited, but rather the means to generate the anatomy. The real target of heritable genetic change is the development by which the trait is produced... in seeking to explain anatomical change in evolution, biologists have come to understand that what they must explain is the changes in developmental processes.
p.179 All embryos start from a single cell... How does a cell... generate the complex anatomy of the embryo and adult? ...the process of development has been called a process of self-organization.
p.179 Cells... express a subset of the organism's genes and cellular behaviors... If the location and combinatorial expression of these processes can be changed in a facile manner, evolution too can proceed in a facile manner.
p.182 Compartments are not present in the egg... Nevertheless, compartments arise by processes inherent in the egg operating under conditions set by the surrounding environment. It is a bootstrapping process where a few small initial differences are acted on to make further differences.
[JLJ - Perhaps for game theory, this concept can explain how a move in a game is 'selected' - a bootstrapping process is applied to a few small initial differences, which make further differences when acted upon. The attention-driving 'compartments' of the 'search' process of an intelligent agent may 'spontaneously' arise from bootstrapping actions acting on whatever little information is present in the richly detailed cues extracted from the environment.]
p.183 We call the compartment plan an "invisible anatomy" because the compartments are only identifiable if one can establish which genes are expressed there... The compartment boundaries, though curiously arbitrary with regard to final anatomy, nevertheless divide the embryo into regions... Stated in terms of the conserved core processes, the compartment map makes possible the use of different combinations of processes at different places in the body.
p.184 compartments serve to make the animal more robust to the environment... compartments are a platform for simple modification in evolution... Compartments were a surprise of such unexpected generality that in the space of ten years, the entire field of developmental biology was refounded on completely different principles.
p.199 Compartments appear in the embryo only at a middle stage of development... The middle stage of development when the compartment map is first present is called the phylotypic stage. It is when embryos of all the different classes of a particular phylum of animals look most alike.
p.209 The compartment map of the phylotypic stage may be special in requiring rather little information for its development. Only a few signals may be needed to orient and scale several of the compartments. The remainder of the complex array may organize itself from these minimal inputs, just as correct placement of a few key pieces of a jigsaw puzzle enables rapid placement of the pieces around them.
p.211 The conclusion of the modeling study, when stated as a hypothesis about real development, is that compartments are relatively easy to set up because they are so adept at completing their activation and self-maintenance... You do not have to turn the key "just right" to get this motor going; any kick will do.
p.213 The robustness of the compartment body plan and its connections by weak linkage to the conserved core developmental processes are key to the facilitation of variety around it.
p.215 Compartmentation as a strategy is one way of generating complexity from a relatively small number of genes and of avoiding conflicts due to use of the same gene product in more than one context.
p.217-218 the developmental circuits, supported by weak linkage and the exploratory mechanisms that make up the body plan, constitute a system that is tolerant of change in the processes that precede and emplace the circuit of selector genes at the phylotypic stage and is permissive of modification that follows the phylotypic stage.
p.219 Chapter 7: Facilitated Variation
p.219 evolvability... gains meaning when resolved into a variation component and a selection component.
p.220 It is the variation component that has been the focus of this book.
p.220-224 We now outline the intact theory of facilitated variation:
- Despite the randomness of mutation... phenotypic variation cannot be random because it involves modification of what already exists.
- The existing organism constrains and deconstrains variation of its phenotype, both the kind and amount... The overall trade-off is such that phenotypic variation is accelerated over what would occur if deconstraint were absent.
- Variation from this trade-off is both less lethal and more appropriate to selective conditions than would be variation from random change. Evolutionary change is thereby facilitated.
- The constrained parts of the organism are the conserved core processes... Their function is to generate the phenotype from the genotype...
- The core processes have been remarkably unchanging over time...
- Most evolutionary change in the metazoa [JLJ - the animal kingdom] since the Cambrian has come not from changes of the core processes themselves or from new processes, but from regulatory changes affecting the deployment of the core processes... Because of these regulatory changes, the core processes are used in new combinations and amounts at new times and places...
- Protein evolution... is itself an example of conserved core processes at work...
- Physiological processes that adapt the individual to environmental conditions are rich targets for evolutionary modification. These processes contain combinations of conserved core processes...
- We propose that a much richer source of targets is to be found in the conserved core processes of development and cell behavior, the processes directed inside the organism rather than toward the environment... Such regulatory modification of existing processes is likely to be less lethal and generate more phenotypic change, for an input of random mutation, than would be gained by inventing new structures or physiology.
- The core processes are built in special ways to allow them to be easily linked together in new combinations, and to be used at new times and places, to generate new phenotypes. These special properties include: a. Weak linkage... b. Exploratory behavior... c. Compartmentation...
- The generation of variation is facilitated principally by: (a) reducing ... the lethality of mutations in one part of the phenotype that might have selectable benefits in another part, (b) increasing the amount of phenotypic change gained for a given amount of mutational change... and (c) increasing the genetic diversity in the population by suppressing lethality.
- Our theory of facilitated variation, stated here in its most coherent and complete form, gives what we think is a plausible account of the dependence of phenotypic variation on genotypic variation, indicating that novelty mostly draws on what is already present in the phenotype, and further indicating the role of conserved components and processes in innovation...
p.226 We suggest that when these conserved processes accumulated in evolution and gained their properties of robustness and flexibility, the organism became more and more a system capable of responding to random mutation and other forms of genetic variation. It did so by using existing processes to produce phenotypic variation, via regulatory changes.
p.226 Instead of a brittle system, where every genetic change is either lethal or produces a rare improvement in fitness, we have a system where many genetic changes are tolerated with small phenotypic consequences, and where others may have selective advantages, but are also tolerated because physiological adaptability suppresses lethality.
p.226-227 In the generation of phenotypic variation from random mutation, the organism as a whole is not a blank slate but a poised response system, rather like one of the signal-response systems within its physiology. It responds to mutation by making changes it is largely prepared in advance to make. Its adaptive envelope of responses is far greater than that which can be elicited by testing every environmental condition on the whole organism... Genetic variation or mutation does not have to be creative; it only needs to trigger the creativity built into the conserved mechanisms... From new combinations we derive whole new physiologies and anatomies, and from state selection we also obtain new physiologies and anatomies, ones that have already been tested. Novelty comes from these two sides and from their interaction.
p.240-241 the power of conserved processes - they can be organized by different means to a similar end. In convergence, similar outcomes are evolved in different ways, making use of exploratory processes, modularity, flexibility, and weak linkage.
p.241 The greater the robustness and adaptability of core processes, the greater the tolerated random genetic variation, which is not eliminated by lethal effects and lessened reproduction.
p.242 In select... systems... we can expect more and more experimental tests of how conserved processes are deployed and more and more evidence of the preexisting poised processes that are evoked by small mutational changes. Facilitated variation will assuredly be exposed to further tests and refinements in the near future.
p.242-243 can evolution be imagined without facilitated variation? ...If animals did not use and reuse conserved processes, they would, we think have to evolve by way of total novelty - completely new components, processes, development, and functions for each new trait. Under these circumstances the demands for "creative mutation" would be extremely high... During the last half-billion years, the anatomical and physiological evolution of multicellular animals has not depended on total novelty, according to what we can ascertain from the fossil record and from comparisons of existing organisms... we have to admit that the conserved processes themselves had to evolve at some prior time, as did their special properties. Facilitated variation assumes the availability of these processes. The evolution of these processes and properties would seem to be the primary events of evolution, requiring high novelty.
p.243 As noted in Chapter 3, the unique and episodic appearance of these processes with the emergence of eukaryotic cells, multicellular animals, and perhaps the first prokaryotic cells, may attest to the rarity of their invention. Once the conserved processes were available, though, the possibility of variation by regulatory shuffling and gating of these processes was unleashed, and shuffling and gating were much simpler than inventing the processes.
p.243 The main accomplishment of the theory of facilitated variation is to see the organism as playing a central part in determining the nature and degree of variation, thus giving selection more abundant viable variation on which to act.
p.243 It is the capacity of the core processes to support variation that we see as the main factor in generating phenotypic variation and in minimizing the lethality of phenotypic variation. It is the nature of these processes, which are poised to generate physiological variation within the organism, that allows genetic variation to be so effective in generating phenotypic variation on which selection acts.
p.243 We think that the organism is so constituted that its own random genetic variation can evoke complex phenotypic change. However, it is the extraordinary power of the conserved core processes that is most responsible for the copious amount of phenotypic variation in response to mutation.
p.244 Chapter 8: Is Life Plausible?
p.247 With facilitated variation, the tripartite Darwinian theory consisting of genetic variation, phenotypic variation, and selection becomes much more complete.
p.248-249 Was evolvability selected? ...Several arguments, taken together, show that the means of facilitated variation should be under positive selection.`
p.260-261 does facilitated variation have something useful to contribute to understanding complex social or political organizations, or elements of design in engineering, or computer science? [JLJ - or game theory?] ...That a comparison of different complex systems might yield new insights for both is the promise that urges a tentative and highly qualified exploration of these areas.
p.262 Perhaps designers of future computers or institutions [JLJ - or game theory investigators] will intentionally borrow features from facilitated variation.
p.264 Perhaps the most important lesson of this analysis may be that the generation of phenotypic variation from genetic variation cannot be taken for granted in complex organisms: it does not reflect random breakdown of the system, but rather a selected design mode of the conserved core processes.
p.275 Glossary
p.293 Notes
p.303 Index
|
