Causal complexity in life

Evolution is the process that generates the relationships between genomes and traits in organisms.  Although we have written extensively and repeatedly about the issues raised by causal complexity,  we were led to write this post by a recent paper, in the 21 October 2016 issue of Science, which discusses molecular pathways to hemoglobin (Hb) gene function.  Although one might expect this to be rather simple and genomically direct, it is in fact complex and there are many different ways to achieve comparable function.

The authors, C Nataragan et al.,  looked at the genetic basis of adaptation to habitats at different altitude, focusing on genes coding for Hb molecules, that transport oxygen in the blood to provide the body's tissues with this vital fuel.  As a basic aspect of our atmosphere, oxygen concentrations differ at different altitudes, being low in mountainous regions compared to lowlands.  Species must somehow adapt to their localities, and at least one way to to this is for oxygen transport efficiency mechanisms to differ at different elevations.  Bird species have moved into and among these various environments on many independent occasions.

The affinity of Hb molecules for, that is, ability to bind oxygen, depends on their amino acid sequence, and the authors found that this varies by altitude.  The efficiency is similar among species at similar altitudes, even if due to independent population expansions. But when they looked at the Hb coding sequences in different species, they found a variety of species-specific changes.  That is, there are multiple ways to achieve similar function, so that parallel evolution at the functional level, which is what Nature detects, is achieved by many different mutational pathways.  In that sense, while an adaptation can be predicted, a specific genetic reason cannot be.

The authors looked only at coding regions, but of course evolution also involves regulatory sequences (among other functional regions in DNA), so there is every reason to expect that there is even more complexity to the adaptive paths taken.

Important specific documentation....but not conceptually new, though unappreciated
The authors also looked at what they call 'resurrected ancestral' proteins, by experimentally testing the efficacy of some specific Hb mutations, and they found that genomic background made a major difference in how, or whether, a specific change would affect oxygen binding.  This shows that evolution is contingent on local conditions, and that a given genomic change depends on the genomic background.  The ad hoc, locally contingent nature of evolution is (or should be) a central aspect of evolutionary world views, but there is a widespread tendency to think in classical Mendelian terms, of a gene for this and a gene for that, so that one would expect similar results in similar, if independent areas or contexts.  This is a common, if often tacit, view underlying much of genome mapping to find genes 'for' some human trait, like important diseases.  But it is quite misleading, or more accurately, is very wrong.

In 2008 we wrote about this in Genetics, as we've done before and since here on MT and in other papers.  In the 2008 article we used the following image to suggest metaphorically the nature of this complex causation, with its alternative pathways and the like, where the 'trait' is the amount of water passing New Orleans on the Mississippi River.  The figure suggests how difficult it would be to determine 'the' causal source of the water, how many different ways there are to get the same river level.

Drainage complexity as a metaphor for genomic causal complexity.  Map by Richard Weiss and ArcInfo

One can go even further, and note that this is exactly the kind of findings that are to be expected from and documented by the huge list of association studies done of human traits.  These typically find a great many genome regions whose variation contributes to the trait, usually each with a small individual effect, and mainly at low frequency in the population.  That means that individuals with similar trait values (say, diabetes, obesity, tall, or short stature, etc.) have different genotypes, that overlap in incomplete and individually unique ways.

We have written about aspects of this aspect of life, in what we called evolution by phenotype, in various places.  Nature screens on traits directly and only on genes very indirectly in most situations in complex organisms.  This means that many genotypes yield the same phenotype, and these will be equivalent in the face of natural selection and will experience genetic drift among them even in the fact of natural selection, again because selection screens the phenotype.  This is the process we called phenogenetic drift.  These papers were not 'discoveries' of ours but just statements of what is pretty obvious even if inconvenient for those seeking simple genetic causation.

The Science paper on altitude adaptation shows this by stereotypical sequences from one individual each from a variety of different species, rather than different individuals within each species, but that one can expect must also exist.  The point is that a priori prediction of how hemoglobin adaptation will occur is problematic, except that each species must have some adaptation to available oxygen.  Parallel phenotype evolution need not be matched by parallel genotypic evolution because selection 'sees' phenotypes and doesn't 'care' about how they are achieved.

The reason for this complexity is simple: it is that this is how evolution working via phenotypes rather than genotypes molds the genetic aspects of causation.

Darwin the Newtonian. Part V. A spectrum, not a dogma

Our previous installments on genetic drift (a form of chance) vs natural selection (a deterministic force-like phenomenon) and the degree to which evolution is due to each (part 1 here) lead to a few questions that we thought we'd address to end this series.

First, there is no sense in which we are suggesting that complex traits arise out of nowhere, by 'chance' alone.  There is no sense in which we are suggesting that screening for viability or utility does not occur as a regular part of evolution.  But we are asking what the nature of that screening is, and what a basically deterministic, Newtonian view of natural selection, that is we believe widely if often tacitly held, implies and how accurate it may be.

It's also important here to point out something that is obvious.  The dynamics of evolution from both trait and genome level comprise a spectrum of processes, not a single one that should be taken as dogma.  A spectrum means that there is a range of relative roles of what can be viewed as determinism and chance that the two are not as distinct as may seem, and that even identifying, much less proving what is going on in a given situation is often dicey.  Some instances of strong selection, like some of chance seem reasonably clear and those concepts are apt.  But much, perhaps most, of evolution is a more subtle mix of phenomena and that is what we are concerned with.

Secondly, we have discussed our view of natural selection before, in various ways.  In particular, we cite our series on what we called the 'mythology' of selection, a term we used to be provocative in the sense of hopefully stimulating readers to think about what many seem to take for granted.  Yes, we're repeating ourselves some, but think the issues are important and our ideas haven't been refuted in any serious way so we think they're worth repeating.

A friend and former collaborator took exception to our assumption that people still believe that what we see today is what was the case in the past.  He felt we were setting up a straw man. The answer is somewhat subjective, but we believe that if you read many, many descriptions of current function and their evolution, you'll see that they are often if not usually just equated de facto with being 'adaptations', and that means that doing what they do now came about because it was favored by the force of selection in the past.  We think it's not a straw man at all, but a description of what is being said by many people much of the time: very superficial, dogmatic assumptions both of determinative selection and that we can infer the functional reason.

Of course everyone acknowledges that earlier states had their own functions and today's came from earlier, and that functions change (bat wings used to be forelegs, e.g.), but the idea is that bat flight is here because the way bats fly was selected for.  One common metaphor going back to an article by Lewontin and Gould is that evolution works via 'spandrels', traits evolved for one purpose or incidentally part of some adaptation, that are then usable by evolution to serve some new function. Yes, evolution works through changing traits, but how often are they 'steps' in this sense or is the process more like a rather erratic escalator, if we need a metaphor?

There are ways for adaptive traits to arise that have nothing to do with Darwinian competition for limited resources, and are perfectly compatible with a materialist view.  Organismal selection occurs when organisms who 'like' a particular part of their environment, tend to hang out there.  They'll meet and mate with others who are there as well.  If the choice has to do with their traits--ability to function at high altitude, or whatever--then over time this trait will become more common in this niche compared to their peers elsewhere, and eventually mating barriers may arise, and a new species with what appears to be a selected adaptation. But no differential reproduction is required--no natural selection.  It's natural assortment instead.

All aspects of our structure and function depend on interaction among molecules.  If two molecules must interact for some function to occur, then mutant versions may not serve that purpose and the organism may perish. This would seem most important during embryonic development.  An individual with incompatible molecular interactions (due to genetic mutation) would simply not survive.  This leaves the population with those whose molecules do interact, but there is no competition involved--no natural selection.  It's natural screening instead.

Natural selection of the good ol' Darwinian kind can occur, leading to complex adaptations in just the way Darwin said 150+ years ago.  But if the trait is the result of very many genes, the individual variants that contribute may be invisible to selection, and hence come and go essentially by chance. This is what we have called phenogenetic drift.  Do you doubt that?  If so, then why is it that most complex traits that are mapped can take on similar values in individuals with very different genotypes?  This is, if anything, the main bottom line finding of countless very large and extensive mapping studies, in humans and even bacteria.  This is basically what Andreas Wagner's work, that we referred to earlier in the series, is about.   It rather obviously implies that which of equivalent variants proliferates is the result of chance.  There's nothing non-Darwinian about this.  It's just what you'd expect instead.

We'd expect this because the many factors with which any species must deal will challenge each of its biological systems. That means many screening factors (better we think than calling them selection 'pressures' as would usually be done).  Most of these are affected by multiple genes.  Genes vary within a population.  If any given factor's effects were too strong, it would threaten the species' existence.  At least, most must be relatively weak at any given time, even if persisting over very long time periods.  Multiple traits, multiple contributing genes in this situation means that relative to any one trait or gene, the screening must be rather weak.  That in turn means that chance affects which variant proliferates.  There's nothing non-Darwinian about this.  It's essentially why he stressed the glacial slowness of evolution.

There is, however, the obvious fact that known functional parts of DNA are far less variable than regions with no known function.  This can be, and usually is assumed to be, the expected evidence of Darwinian natural selection.  But factors like organismal dispersion or functional (embryonic) adequacy can account for at least some of this.  Longer-standing genes and genetic systems would be expected to be more entrenched because they can acquire fewer differences before they won't work with other elements in the organism.  This is at least compatible with the view we've expressed, and there could be some ways of testing the explanation.

This view means we need not worry about whether a variant is 'truly' neutral in the face of environmental screening.  We could even agree that there's no testable sense in which a variant evolves by 'pure' chance. Even very tiny differences in real function can evolve in a way that is statistically 'neutral'.  Again, this can be the case even if the trait to which such variants contribute is subject to clear natural or other forms of selection.

This view is also wholly compatible with the findings of GWAS, the evidence that every trait is affected by genetic variation to some extent, the fact that organisms are adapted to their environment in many ways and the fact that prediction based on genotyping is often a problematic false promise.  And because this is a spectrum, randomly generated by mutation, some variants and or traits they affect will be very harmful or helpful--and will look like strong, force-like natural selection.  These variants and traits led to Mendel, and led to the default if often tacit assumption that natural selection is the force that explains everything in life.

Further, it is important for all the same sorts of reasons that the shape of the spectrum--the relative amount of a given level of complexity--is not based on any distribution we know of and hence is not predictable, generally because it is the result of a long history of random and local context and contingencies, of various unknown strength and frequency (about the past, we can estimate a distribution but that doesn't mean we understand any real underlying probabilistic process that caused what we see).  This is interesting, because many aspects of genetic variation (and of the tree of life) can be fitted to a reasonable extent to various probability distributions (see Gene Koonin's paper or his book The Logic of Chance).  But these really aren't causal parametric 'laws' in the usual sense, but descriptions after the fact without rigorous causal characteristics.  Generally, prediction of the future will be weak and problematic.

In the view of life we've presented, evolution will have characteristics that are weak or unpredictable directional tendencies, and the same for genetic specificities (and hence predictive power). It is the trait that is in a sense predictable, not the effects of individual genes.

We think this view of evolution is compatible with the observed facts but not with many of the simplified ideas that are driving life sciences at present.

Our viewpoint is that the swarm of factors environmental and genomic means that chance is a major component even of functional adaptations, in the biodesic paths of life.

Phenogenetic drift

Many people think that biological traits are due to specific genes and that variation in a trait is due to variation in that gene.  So it would follow that if a gene variant becomes more frequent, the trait variant it codes for becomes more frequent.  We get this idea from Mendel, but it explains only a small fraction of traits; most traits are due to many genes, and this fundamentally changes the specificity of the genotype phenotype connection.  To complicate this further, a house may be made of bricks, but the bricks can all be swapped for different bricks, and you've still got the same house.  This applies equally to genes and traits.

It may sound mystical to suggest that biology is not "molecular'' at its core the way physics and chemistry are.  How can it not be?  Life consists of molecules undergoing biochemical reactions that must follow physical laws.  We think we understand those laws, but if not, and someone were to discover that they needed revising, life would follow whatever revisions to those laws we came up with.  But no one seems to think that currently.  Which is not to imply that our understanding of how life works--how those molecular principles apply--is necessarily correct.

In particular, genomes are molecular entities, and the prevailing theory about life is that genomes and their variation drive life and its properties and variation.  Under a rather shallow interpretation of Darwinism, genomes bearing specific genotypes that succeed by proliferating themselves because of their functional success, and hence their specific sequence details, are represented with increasing frequency over time.

But suppose it is not a genome per se that is especially conserved by evolution. Suppose the trait, the ephemeral phenotype that is refreshed each generation by new embryos and persistent over time, is really what we need to understand.  A phenotype, an individual, is an 'emergent' result of genotypes that is, at present at least, only very imperfectly predictable from its genotype.  Since we know that similar phenotypes can be generated by a variety of genotypes, individual genes would then be "only'' the meandering spoor left by the process of evolution by phenotype.  Over time a very different set of genotypes might generate a favored phenotype, compared to the genotypes that did so at some time in the past.  This phenomenon is called phenogenetic drift.

Perhaps biology has hidden behind the Modern Synthesis, and the idea that all the action is in gene frequencies, for too long. Life is ultimately about phenotypes, the result of interactions by large numbers of genes and other molecular factors, and a better theoretical basis for understanding the dual evolution of phenotype and genotype--the tempo and mode of phenogenetic drift--is needed.

Biology struggled for much of this century to achieve respectability in the pantheon of science, and by mid-century found its "atoms'' in genes. If atoms are 'it', biology had it made!  But evolution is less specific and determinative than prevailing, if elegant, molecular and evolutionary theory suggest. Evolution works by phenotypes, whole organisms that reproduce or don't, not genotypes. A phenotype may a vague and ephemeral notion that is a poor excuse for an atom, but it may be the basic "unit" of biology nonetheless, and one we should strive to understand, on its own terms, with the many new methods that now exist.

This is yet another reason to be more circumspect about genomic determinism than many currently are.  It is why the determinism of this person's phenotype by his/her genotype may be unique.  It's why Darwin's necessary focus on the whole organism rather than just its 'gemmules' was insightful, despite his totally wrong theory of inheritance, and even in an era of rapid and fashionable excitement about the nature of molecules, and specifically genes, as the fundamental physical particles in Nature.

There are probabilistic aspects to gene action, but it may be more important to realize that all genomic functional units are susceptible to variation through mutation, and it is the interaction of countless such units that generates traits, many of whose actual values depend on the environmental factors that interact with DNA's and its products. It is far easier to think of genomes as consisting of beads on a string, than of them as 'mere' contributors to emergent interactions, but the latter is closer to the truth, as countless experiments have by now clearly shown.  But, if there is such a thing as a good theory of 'emergence', we don't yet have it, and clearly don't know how to apply it to the eons of ad hoc events, mutational and selectional, that have generated what is here today, not to mention the same sorts of 'slippage' that intervenes between genomes--a person's born DNA sequence, and the person's traits.

Phenogenetic drift, which includes the undeniable equivalence among many different genotypes in terms of the trait values with which they are associated, shows why when we use reductionism to dig below the level of the organism or its traits, to its genome, we pass through the very organizational phenomenon we are trying to understand.

On the mythology of natural selection: Part VII. Phenogenetic drift

We are trying in this series of posts (see this week and last, beginning with Part I here) to enumerate a few ways in which organized traits can arise without the usual canonical view of natural selection as the complete, force-like causal process.  We do this to temper what we see as an often unquestioning belief in natural selection.  We do this because such a belief leads to what we think is extreme, and unwarranted, genomic determinism and belief in inborn inherency that leads both to potentially disastrous societal consequences and false hopes in regard to promises for health and well-being.

Also, mistaken reliance on simplistic explanations impedes the effort to understand the truth rather than mythology about the truth--because mythology removes a sense that we must work harder to understand the actual truth.

Today, we wish to note an important aspect of the relationship between genes and traits, even if one were to accept a purely "Newtonian" law- or force-like version of ubiquitous natural selection as the, that is the, cause of traits in organisms.

The point is that even then, and even if genes (broadly defined, without quibbling about what a 'gene' is) worked in a perfectly force-like deterministic causal way, even ignoring all the obvious probabilistic aspects or environmental components, even then, genetic determinism is not the simple story as is often portrayed.

"Survival of the fittest"?  A non-sensical notion too widely adopted
It is blatantly clear to anyone wishing to make even the most casual observation of nature rather than a book of slogans, that 'survival of the fittest' is at the very best a misleading term.  What does 'the' mean? The one and only most-fit individual?  Clearly not!  All those who are fit (here 'fit' refers to success in the evolutionary screen of natural selection)?  Or somewhat fit?  If that, then what does 'fittest' mean?  All those 'equally' best or only the very best survive?  That is an untestable definition.

Darwin should not have adopted this phrase, which he borrowed from Herbert Spencer to clarify, one might say, his term 'natural selection' to make sure that no one would think he (Darwin) was imagining God as nature's selector. The plain and manifest truth is that 'survival' is not exactly the right term so Darwin mis-spoke or spoke metaphorically, because it is both survival and reproduction that are important, and for most species not just survival per se but length or timing of survival, etc.  What Darwin probably meant was 'survival' in the sense of being represented in the next generation.  In any case, semantics aside, in the hurly-burly of real life, evolutionary success is a problematic, quantitative rather than simple qualitative yes-no phenomenon.

It is common if not typical or even necessary that biological traits are produced by the action of many functional elements of an organism's genome, not just one.  Traits themselves usually have at least some variation among individuals within and between species (and during each one's life).  Here we ignore environmental factors, but their variation is of course often an important additional contributor to trait variation. Essentially, life is causally a many-to-many phenomenon.

This fact has profound implications for our understanding of phenogenetic relationships, that is, relationships between genes and the traits to which they contribute.  To show this clearly, in what follows, we will for the sake of argument just assume the force-like universal view of simplistic natural selection.

Divergence of primordial EMP (enamel) gene. Phenogenetic drift. Source: Kawasaki and Weiss, 2003

Phenogenetic drift
With traits that are affected by many different genes (often called 'polygenic' as a short-hand term), many different genotypes can yield essentially the same phenotype.  This is we have called 'phenogenetic equivalence'.  In the Darwinian arena, individuals with the same trait will have similar fitness prospects--they'll be treated similarly by natural selection--even if their trait is due to different genotypes.  That means that the contributing genetic variants are equally 'fit': they proliferate equally well. Different individuals in a population, or individuals from different populations, or individuals from the same population over different time periods, will have the same traits for different genomic reasons. This kind of causation is essentially the definition or essence of polygenic or causally complex traits.

When this occurs, along with the chance elements in life itself, the chance elements in recombination among genes in genomes and between parents and the gametes they provide to each offspring, the relative frequency of the contributing genomic variants will vary over place and time essentially by chance--they will drift as the term has it. This is the case even when selection of the classical kind is at work, even when the selection is strong. This is phenogenetic drift, or chance changes in the relationship between phenotypes and genotypes. (The phenomenon was discussed in the reference below***, and elsewhere in my work, and see Kawasaki on SCPP genes and mineralization, or Wagner on avian digits, e.g.)

With phenogenetic drift there is no reason to expect that a given gene or genetic variant is necessary or sufficient for the trait.  Genetic determinism has a different kind of meaning than the usual 'marginal' (statistically, on-its-own) view of genetic causation.  One could say that the Predictance, the probability of a given phenotype for a given genotype, was very high, but the Detectance, the probability of a given genotype underlying a given phenotype, was low.

But the common reality, based on countless GWAS and other types of genomewide enumeration studies to relate phenotypes to genotypes, is that such prediction is usually small, trivially so for each individual variant and even if all statistically detected genetic variants are taken into account.  Now and then a strong-effect variant at a specific gene is identified, and one might find evidence that fitness--health or actual survival--is predictable from the genotype at that specific gene.  But that is the exception, the Mendelian tease, the first taste of a drug that leads to the hyper-Darwinian addiction.

The Mendelian tease: peas that followed rules; Ernst Benery Erfurt, 1867.

In sum
For rhetorical purposes we have assumed here that the world is a deterministic Darwinian one, but in fact environments are at least as complex as genomes, the interactions among genes are complex, and probabilistic elements are involved all along the way.  We cannot escape a certain amount of probabilism, either because that is the true essence of biological causation, or at least because our measurements are imperfect. Worse, we don't know all the factors to measure, and when it comes to environments they are always changing and in directions we simply cannot predict.  This is particularly true in the case of humans, because our behavior is based on all sorts of unpredictable cultural elements, so that, for example, disease risks are inherently estimated from past exposures, and our future exposures (diet, environmental chemicals, climate, etc. simply cannot be predicted).  All these factors introduce slippage between genotype and phenotype at any given time, and hence over evolutionary time.

Again: the bottom line is that when many genetic factors contribute to a trait's variation, the combination underlying any given individual's trait can be unique to that individual.  It can be problematic to predict the trait from the genotype (as in 'personalized genomic medicine') or to predict the underlying genotype of an individual's phenotype.  Genetic causation is typically not as deterministic as its widespread, if often implicit, image.

Phenogenetic drift is an obvious fact of life, and it raises important questions related to the DNA sequence conservation issue we considered in the context of functional selection earlier in this series. That's because when contributing factors are experiencing phenogenetic drift, specific genes or variants need not be particularly conserved.  So how is it that when phenogenetic drift is part of life and evolution, there is so much evidence at the gene-by-gene sequence level, for purifying selection, for sequence conservation?  Here is a serious subject for study, though it poses no sort of controversy about adaptive evolution except by showing why simplified views of natural selection are inaccurate and at best incomplete.

The so-called Modern Evolutionary Synthesis, formulated in the 1930s and 40s, united paleontology, Darwinian gradualism, and Mendelian inheritance into a single gene-based view of life and its evolution.  It was, essentially and at least implicitly, focused on the effects of variants at single genes, screened by natural selection.  The theory of population genetics was its mathematical basis, and is usually presented for simplicity's sake in textbooks and classes as focused on single 'Mendelian' (two-allele) models, just like green and yellow peas.  But this almost cartoon-like simplification has widely been implicitly or or even explicitly accepted as the reality, even in current medical school curricula (and widely in 'gene-for' research, a topic we often write about). This view in practice often treats individual genes as having inherent deterministic (causal) value, on their own, free of much recognition of context.

Phenogenetic drift, like other not-Darwinian aspects of genotype-phenotype relationships and their evolution, is simply observable, not mystical, perverse, or in any way arcane or secret.  It belongs in the panoply of tools we have to attempt to understand biological causation and its evolution, fleshing out the skeleton of the process that Darwin was able to intuit with the tools available in his time, and as a corrective to the caricature-like simplism that is so widespread today, even in many professional circles and in the public media.  Of course, nobody admits to being simplistic--but pay attention to what they actually say and how they say it, to see whether you think our impressions are accurate nor not.



***Weiss, K, Fullerton, SM  Phenogenetic Drift and the evolution of genotype-phenotype relationships.  Theoretical  Population Biology, 57: 187-195, 2000.

Dogs, crows and garbage: phenogenetic drift

Scavenger hunts
Monday's the day trash gets picked up in our neighborhood. When I was out running early this Monday morning, I passed a scene I see not infrequently -- a group of crows congregated around a plastic trash bag, its contents spilling into the street and the birds tugging at anything in it that looked appetizing.  As I ran by, two of the birds flew but one brazen bird stayed right where he was, guessing that I was no threat to him.

Crows are very smart, and not just because they've figured out how easily plastic garbage bags are breached.  I've also seen them congregated around trash barrels with the lids knocked off.  I've never seen them actually take the lids off, but it must be they do.  When I was a child it was dogs who got into the trash.  This was before leash laws, when it really was a dog's life.  Dogs we keep as pets now have only ancestral memory to recollect the halcyon days of wandering the streets untethered on trash day.  Now it's a crow's life.

But sometimes it's bears, too.  A mother bear and her three cubs were making the rounds of our neighborhood a few weeks ago, knocking down bird feeders and compost bins, and this isn't an unusual occurrence.  The wilding of America?

This morning's sighting reminded me of a piece in the February 21 New York Review of Books by Russell Baker in which he reviews a book by Jim Sterba, Nature Wars: The Incredible Story of How Wildlife Comebacks Turned Backyards into Battlegrounds. I have not read the book, though now that I'm remembering this, I just might have to. Unhappily for the author, Baker's is one of those pieces that makes you think can skip the book, even as he gives it a fine review.

In any case, Baker describes seeing a pair of foxes mating in his garden situated, he says, just two blocks from the county courthouse in the bustling center of town. The encroachment into human-populated areas of animals who were once sighted only at a distance is becoming increasingly more common.

Sterba ... argues persuasively that events like this foxes-in-the-garden sighting are evidence that humans are losing some kind of property rights struggle with creatures of the wild. He cites an extensive history of resolute and sometimes blatantly hostile real-estate invasion by beavers, Canada geese, wild turkeys, and white-tailed deer, all of which were once assumed to be picturesque and even lovable denizens of the dark and safely remote forest. In-town appearances by coyotes and bears are now commonplace in communities across the country, and trespassers in my own garden, aside from the foxes, have included groundhogs, possums, skunks, feral cats, and one blue heron that ate all the koi with which we thought to beautify the fish pond.

Lucky heron.  As Baker points out, in the last fifty years these animals have discovered that life can be a lot easier closer to town, and food a lot more abundant.

The woods have no garbage cans and dumpsters filled with discarded food, no lovingly tended tomato plants, no ready-to-pluck dahlias and nasturtiums, no tasty, newly planted shrubs. 

Best of all from the animal viewpoint, humans are no longer the same dangerous predators who once pushed the beaver close to extinction and reduced the entire North American white-tailed deer population to a trifling 500,000 scarcely a century ago [there are now 25-40 million]. Sterba believes that this human withdrawal from combative relations with woodland animals is one of the major causes of their proliferation: man as killer has undergone a softening change.

Indeed, the crow standing in the middle of the road waiting for me to run by its food source is evidence of exactly that.

It doesn't matter who does it
But here's the thing -- whoever's doing the scavenging, there's still garbage spread all over the street.  And this is an apt metaphor for an important way that evolution works.  Yes, gene function is often conserved, so that the same gene (Pax6) is involved in photoreception in the eye of fruit flies and frogs and humans, and the human form of the gene can be transplanted into even distantly related species and the eye will still be made.  But it's also true that often a trait is conserved but the genetic scaffolding is very different.  Natural selection might preserve the phenotype, the trait, but it can't see the genotype, leaving it free to vary.

Ken and his then post-doc, Malia Fullerton, published a paper in Theoretical Population Biology in 2000 describing just this, though others have described it as well (reviewed by Brian Hall in his 2003 book, Keywords And Concepts In Evolutionary Developmental Biology.) The effect is called 'phenogenetic drift' to indicate that the trait's genetic basis can change. This is not the same as genetic drift, which is when genetic variation that has no effect on a trait, or at least on reproductive success ('fitness'), changes over the generations, and it's not the same as phenotypic drift, when traits vary over generations but that variation doesn't affect fitness.  This happens, therefore, regardless of whether there is selection, even strong selection, or not.  A prerequisite, or partial one at least, is that many genes contribute to the trait, so that different combinations of variants in these genes can lead to similar traits.

Hall uses the example of proteins that make up the lens of the eye. They can be unrelated among taxa, as long as they let light pass through. Kazu Kawasaki, a very skilled research scientist in our lab, has written a number of papers describing the evolution of mineralization in vertebrates. One of his early papers is on the changing genetic basis of vertebrate teeth. He has found a gene family, that he calls SCPP genes, that varies widely among species, and yet contributes to the formation of mineralized tissues -- teeth and bone -- in all of them.

As Hall concludes his section on this subject, "Phenogenetic relationships are less determinative for more complex traits."  There are many genetic pathways to being tall, or to developing heart disease, or to being a fast runner  And that's before we even throw environmental factors into the mix.  Genetic determinism, which we've blogged about often, is simply too easy to assume, often incorrectly so.  When you look out the window on trash day and see garbage strewn all over the street, don't leap to judgment about who did.  It's only when you find the bear scat or a crow feather that you'll know.  But even then, do allow for the possibility that the neighbor's dog got loose.

New genes, old function

Ken mused a while back here on MT about the improbability of finding a DNA sequence that had no similarity to sequence from any known organism.  And this is as we'd expect, if all life on Earth shares a common ancestor, and nothing that has been discovered since Darwin first proposed this in 1859 suggests otherwise.

So, why are researchers reporting DNA enzyme sequences that don't appear to be homologous to any known sequences for similar enzymes?  François Delavat et al. have sequenced genes from organisms found in an acid mine drainage, organisms that are resistant to being cultured (i.e., that can't be readily grown up in the lab), looking for novel genes or function.  They report their findings in the open-access Nature journal, Scientific Reports ("Amylases without known homologues discovered in an acid mine drainage: significance and impact").

Amylases are enzymes that catalyze the breakdown of carbohydrates, or in the case of bacteria described here, degrade polymers found in their acidic, metal-heavy surroundings.  Amylases from bacteria that grow in culture have been well-studied, and have been classified into several families based on their structure and other characteristics.  Some have been found in extreme environments, but no one had reported the sequencing of DNA from Acid Mine Drainages before this paper.  These are very low pH, very high metal environments. 

The authors

...decided to perform a function-based screening for the well-known amylases, using standard techniques. This strategy allowed the isolation of 28 positive clones, 2 of them being subcloned, the proteins purified and characterized in vitro. In silico analyses based on the nucleotidic sequence and both the primary and the predicted tertiary structures revealed that they are completely different from other known hydrolases as both genes encode a « protein of unknown function » and display no known conserved amylolytic domain. Nevertheless, in vitro tests confirmed the amylolytic activity of these 2 enzymes.

That is, these genes did degrade polysaccharide, but neither of the subclones matched any known amylase sequences in the databases. 

As Delavat et al. point out, much is known about lab-friendly bacteria, and a whole lot less about organisms that can't be grown in the lab.  Thus, if these results are confirmed, the fact that these genes, from organisms that were found in an extreme previously unexplored environment, don't look like other known amylase sequences doesn't at all suggest that these bacteria are unique, or that they don't share the same common origin the rest of us share.  Rather, it suggests that the lab-centric biology of the last century has given us a lab-centric view of the world.  It's no surprise that bacteria that live in the low pH, high metal extremes of Acid Mine Drainages would have evolved particular enzymes appropriate for that environment.  But it's also not a surprise that these enzymes have a function that is common to bacteria in every environment.

That the genes that code for these enzymes are unlike any of the subset of amylases yet described is another example of phenogenetic drift, the conservation of a biological trait or function even when its underlying genetic basis has changed.  These genes may look novel now, but as more bacteria are characterized from non-lab environments it's likely that more will be found that share some of the characteristics of these amylase genes. 

Note, also, that these sequences are genes -- they have protein coding structures and are identifiable from sequences as such.  They are not 'random' sequences with no known relation to the usual characteristics of genes.

A major evolutionary transition explained

Photo from the BBC website

It's difficult to compete with the Higgs Boson story, but here's a strong second, a hot breaking story on the BBC web site about erections in big birds (note, the picture shows the birds, so you'll know what we're talking about, but the erect structures in the picture are necks).  The story starts out:

Ostriches have bloodless erections, according to researchers.

The large birds were previously thought to have blood-based erection mechanisms similar to humans.

But scientists from Yale University, US, have now confirmed that the birds actually enlarge their penises with lymph fluid.  

All other birds with a penis achieve erections in this way, leading scientists to believe the mechanism evolved in their ancient ancestors.

The paper itself is published in the Journal of Zoology.  The authors explain in the abstract why it's so important:

Because the penis in all other described birds has a lymphatic erection mechanism, clarifying that the erection mechanism of ratites [large flightless birds] is of great importance to understanding one of the major evolutionary transitions of penis morphology within amniotes. Here, we show that the erection mechanism of ratites is lymphatic, confirming that the evolutionary transition to lymphatic erection occurred in the last common ancestor of Aves.

You'd always wondered about the major evolutionary transitions of penis morphology, hadn't you?
(Actually, though, this may not have been as pressing a question as it seemed at the time, given another recent breaking story: men don't think about sex as often as their reputation has lead us all to believe, proven by a study that asked men how often they thought about sex. 

"The absolute number of sexual thoughts is dramatically less than the urban legend that men think about sex every seven seconds," says study researcher Terri Fisher, PhD, professor of psychology at The Ohio State University at Mansfield.

Men were asked to push a button on a counter every time they had a sexual thought.  But, we wonder, isn't it possible that the count was artificially reduced by the subjects' thumbs tiring of pushing the freaking button?) 

But, back to the story at hand, ratite penises.  There was a suggestion long ago of lymphatic involvement in the erections of flightless birds, but it remained unconfirmed until now.  After all, who dared to look?  These guys bite hard!  (Hard to do it, but we're resisting the temptation to make a pecker joke here.)  Instead, the authors were sent ostrich and emu penises from birds of reproductive age (that is, certified to be adults, if you're worried that we're writing this from Penn State), and they dissected them -- for methodological and anatomical details, see the paper, but suffice it to say that their work confirms the old suspicion. 

It is rather surprising that this wasn't known until now, and the authors of the paper briefly describe why this is so, including something tantalizing about a reference gone missing.  Indeed, there are a lot of citations in the paper of "unpublished data" on bird penile morphology.  So, it's even more surprising that all it took to answer this question was a bit of dissection of the relevant organs.  Now done. 

The story on the BBC reports that there are still unanswered evolutionary questions, however (there must be, or the specialists would be out of a job).

Similarities have been drawn in the past between bird and reptile penises but the latter use blood for erections, as do mammals.

"The reason why the change between blood vascular and lymphatic took place remains a mystery," said Dr Brennan.

"The lymphatic system is a low pressure system, so this means that erection cannot be maintained, and this has some important implications for how birds actually copulate," said Dr Brennan.

Some species of bird, such as ducks, are known for their "explosive" erections achieved when lymph fluid is forced into the penis to increase pressure for a short time.

But ornithological reproduction expert Prof Tim Birkhead from the University of Sheffield suggests that the structure of ostrich penises could make up for the shortcomings of the lymph system.

"Ostriches and rheas appear to have additional muscles that help to maintain a rigid phallus," he explained.

Ok, so is this actually of interest, other than as a footnote to the sexual life of birds?  Because, after all, however erections happen in these birds, the mechanism has worked for a very long time (except for the dodo and the roc).  The authors say this is an important contribution to the understanding of the evolution of penis morphology.  We have to take their word for it, though this does seem to be a rather specialized interest -- yet another science paper detailing perfectly good work, but that didn't warrant all this fuss.

Is there even a general lesson here?  Well, it's yet another of the myriad examples of variation underlying a single trait or mechanism -- if it does the job, evolution can support it.  Nothing new here.  Indeed, the genetics underlying many examples of what's called 'phenogenetic drift' -- many genetic pathways to a given trait -- are well-documented (Kazu Kawasaki in our lab has done a lot of work on the evolution of genes for mineralization in multiple lineages, for example, a beautiful example of phenogenetic drift; same trait, even under strong selection, yet produced by different genes in different individuals, populations, or times).  So the fact that there are different erectile mechanisms isn't at all a surprise, just another variant in the wide spectrum of how organisms reproduce.

What makes this story manage to stand tall in the news makes it harder, to fathom.

Does phlebotomy 'work'?

There's a discussion in a nice book about the history of Islamic science (Ehsan Masood, Science and Islam: A History) of a man named al-Razi, who in about AD 900 was said to have done a carefully controlled experiment to test whether phlebotomy (blood-letting) worked as a treatment for meningitis. Some patients were given the treatment and others were untreated 'controls'. Al-Razi found that the bloodletting worked, in that more of the treated patients than controls recovered.

This therapy was part of the ancient and revered view of life upon which the classical medical approach codified by Galen was based. The humoral theory, that existence and hence life and health are based on balance of four basic properties (earth, air, fire, water), that in humans corresponded to blood, black bile, yellow bile, and phlegm.

Everything could be explained in terms of disease as the state in which these are out of balance. Blood-letting was done when the patient was deemed to have an imbalance by an excess of blood. Galenic medicine lasted for many centuries and it was verboten even to question it. And why question it? It worked! That is, some patients got better and the belief in the system led everyone to accept its sometime success as supportive evidence (and indeed, it's possible that even when assessed by modern scientific standards, bloodletting may sometimes have done some good, as this story describes).

Why don't we accept it today? In fact, even al-Razi himself wrote a book casting doubt about the degree to which Galenic medicine was true. After all, we accept modern medicine even though it fails to cure everyone. We have to ask what causation really is. After all, placebos work. If you know people are praying for you, it apparently works -- although prayer doesn't work if you don't know people are praying for you.

We dismiss that as 'only' psychological, even if that is purely physical and molecular, by involving neurotransmitters that affect other cell behavior, such as by the immune system and who knows what else, eventually leading to improvement in the disease. Blood-letting apparently has a measurable, replicable physiological rebound effect that makes people feel better a few hours later. We say these things don't really cure the disease, or if they're just psychosomatic, somehow that doesn't count. But if the brain is a material rather than immaterial structure, and the effect is thus material, why doesn't it count?

We want higher percentages of success. We want therapy to be direct, rather than indirect. If the treatment is believed by the patient, it boosts his immune system in some way, etc. Somehow, targeting the true pathology indirectly, rather than by targeting the proximate molecular cause, is not considered 'real'.

But that's our own culturally derived way to define medicine and its efficacy. It's similar with diseases like AIDS and HIV. As the South Africans said for a decade or more, poverty is the true 'cause' of AIDS, not the virus. Unfortunately, many thousands died as a result. Yet poverty is still causally associated with HIV infection. South Africa has finally accepted that HIV is also a cause of AIDS, and thousands or millions of lives may now be saved as a result.

Empirically, the desired explanation can be chosen to be some net result -- 'cure' in the case of disease. Science in the west, at present, wants reductionist molecular explanations, about proximate cause. Causes higher up the material chain -- like poverty and poor education cause poor neighborhoods with no good grocery stores cause reliance on McFastFood causes obesity causes high blood pressure or glucose causes retinal and peripheral neuropathy causes blindness and loss of extremeties. So what causes blindness? Even in our molecular, reductionist, technical age, diabetics still become blind.

There is no one answer. If removing poverty greatly reduced blindness, isn't poverty a cause? Or McBurgers? The prevailing view is that if we identify some ultimate cause -- the preferred target for many in science these days is your 'personalized' genome -- we will get to the 'real' cause and will then live forever. But the focus on genes is part and parcel of the structure of our current society.

Whether one approach to causation will ever, by itself, lead to miraculously high levels of efficacy nobody can say. Galenic physicians thought they had the ultimate answer. Collinsian medicine (Francis Collins, Director of NIH and the chief spokesperson for personalized genomic medicine) is having its day today. What about tomorrow?

The same kinds of questions arise in evolutionary and developmental biology. We've recently posted on phenogenetic drift-- the idea that essentially the same trait can come to be due to different genetic bases even while being conserved by natural selection -- which suggests that genes contribute but are not 'the' cause of the trait. This is related to the entire concept of complex causation.

So was al-Razi right that phlebotomy cured meningitis? Perhaps it is inappropriate to ask whether Galenic medicine 'works'. It is more interesting, to us at least, to ask what we mean by 'works'.

[p.s., al-Razi, known in the west as Rhazi, wrote critically about Galenic medicine in a book Doubts About Galen]

Plus c'est la même chose, plus ça change??

Nothing ever really changes, or so goes the old French saying 'plus ça change, plus c'est la même chose.' The more things change, the more they're the same thing. But is that how life really is?

An interesting new paper in Genome Research on this subject is getting some notice, but in fact the results should not be surprising, given what is known about evolution and development.

The blurb in Nature says:

The earliest stages of embryo development [as shown in the photo above, taken from the paper] seem to be almost identical among mammals. However, Sheng Zhong at the University of Illinois at Urbana-Champaign and his team have found that 40.2% of the genes shared by humans, mice and cows are expressed differently at this point.

Their analysis of gene-expression patterns in embryos at various stages early in development showed that differences result from altered gene regulation. In some cases, mutations affected the binding of regulatory proteins. In others, transposons or 'jumping genes' had hopped in front of the genes, changing their regulation.

This variation among species suggests that multiple gene networks can guide embryo development, and could be harnessed to generate embryonic stem cells.

This may have significant implications for stem cell research, but it also says a lot about evolution and development (EvoDevo). People are often surprised to learn that traits that are similar between species can have very different genetic architecture -- it's more usual to hear about homologous or orthologous genes, similar genes for similar traits (the iconic example is Hox genes, which play a major role in body patterning in fruit flies and humans, and every species in between).

Alternatively, a story in the New York Times last week described the usefulness of yeast or plants or worms for finding genes for human diseases. The genes described belong to clusters that do completely unrelated things in different organisms (one example was of a cluster of genes that repair damage to the cell wall in yeast but that is involved in blood vessel formation in humans). These clusters have been conserved through evolutionary time.

But the idea that different genes can underlay homologous traits is perhaps more counter-intuitive. Ken and his then post-doc, Malia Fullerton, published a paper in Theoretical Population Biology in 2000 describing just this. The effect is called 'phenogenetic drift' to indicate that the trait's genetic basis, the genetic effects that generate the phenotype, changes. This is not the same as genetic drift, when genetic variation that has no effect on a trait, or at least on reproductive success ('fitness') changes over the generations, nor that of phenotypic drift, when traits vary over generations to the extent that the variation doesn't affect reproductive success. Phenogenetic drift can be these things, but also, and perhaps most importantly, it occurs even when there is selection affecting the trait, even strong selection.

Thus, in the example above, the very early and hence very fundamentally important, stages in mammalian embryos are quite similar among different species, but the usage of genes to make it so is considerably different. Phenogenetic drift is easy to see all over the place -- even within species, when different people can have the same trait -- even including disease -- for different genetic reasons. Relative to each other these genotypes are equivalent in fitness terms, and their contributing alleles will change frequency over time by drift. It's no surprise to see it when we know that duplicate genes and many contributing genes together generate lots of redundancy and alternative genetic pathways to get the same trait.

However the dogma about selection and adaptive evolution has been gene- rather than trait-centered. But when many different genotypes can generate effectively the same phenotype, then perhaps genes really aren't the most important things to consider when we try to understand evolution. In this sense, perhaps because genes weren't understood at the time and perhaps because he had it right, Darwin developed his ideas about phenotypes -- the traits of organisms -- rather than genotypes. After all, he didn't need his idea of genetics to explain the historical nature of life's variation (indeed, evolution as he saw it wouldn't work under his idea of inheritance, and he knew it).

It often seems that the longer a trait is maintained, the more likely it is to have changed its genetic basis. Kazu Kawasaki in our lab, for example, has written a number of papers about the different genetic architecture for bone and tooth mineralization that has evolved in different lineages. In that case, the gene pathways share a common ancestor, a single gene that duplicated in different lineages to create gene families involved in mineralization -- he calls these the SCPP gene family. The original founding gene was apparently involved in the development of the first vertebrate mineralized skeletal tissue (probably, external skeletal protective plates or scales, even before there were calcified bones).

The final composition and structure of mineralized bone may vary, but it's the same trait, and serves essentially the same purpose (strength) in different lineages. Same trait, different genes. Whether this is usually true with conservation of form within lineages (such as the ants in amber that we wrote about last week), we can't know, because we have no way to know the original genetic basis. However, among other things, phenogenetic drift or alternative genomic pathways to the same trait, has interesting implications for notions of homology, the sharing of traits today because they have descended from the same ancestor. Traits can be homologous, but their genetic basis may not be. When selection is on the phenotype, it gets maintained however works!