t f p g+ YouTube icon

Understanding Evolution: An Introduction to Populations and Speciation

Bookmark and Share

September 15, 2011 Tags: History of Life

Today's entry was written by Dennis Venema. You can read more about what we believe here.

Understanding Evolution: An Introduction to Populations and Speciation

One of the challenges for discussing evolution within evangelical Christian circles is that there is widespread confusion about how evolution actually works. In this (intermittent) series, I discuss aspects of evolution that are commonly misunderstood in the Christian community. In this first post, we examine how speciation is something that happens to populations over many generations, and discuss how this informs our understanding of human speciation.

One abiding misconception of evolution among Christians (and even for students of biology) is that the production of new species is a sudden event, or one that begins through a single breeding pair (for sexually-reproducing organisms). In reality, new species arise through incremental changes to populations, not individuals.

Speciation is a population-level phenomenon

Imagine a child’s flip-book, where the image on each page is slightly modified from the one preceding it. Let each page in the book represent one generation for a population of organisms (NOT individuals). Each page in the book is very similar to pages on either side of it: indeed, if you placed adjacent pages side-by-side you would need to study them carefully to find the tiny differences. Pages widely separated in the book would have more obvious differences. So too for populations over time. Each generation, as a population, is slightly different at the genetic level than the one preceding it. These changes need not be new mutations, though each generation does introduce a small number of new mutations into a population. It may be that certain variants of genes (called alleles) have increased or decreased in frequency in the population, or that genetic recombination during sexual reproduction has generated new combinations of previously existing alleles. These subtle differences may slightly change the average characteristics of the population, just like turning a page in the flip-book.

The fact that the average characteristics of a population can slowly change over time creates an interesting possibility: if two separated populations of the same organism accumulate enough differences over a large number of generations, then the resulting populations may become different enough to prevent interbreeding even if the opportunity later arises. All that is needed to start this process is some mechanism to reduce or eliminate the exchange of genetic material between the populations (so-called “gene flow”). This blockage is needed, or the average genetic makeup of the two populations will remain the same. Only when a genetic barrier is in place can genetic differences begin to build up between the two groups. There are different mechanisms by which this occurs in nature, some of which we will examine below.

Separation by geography (allopatric speciation)

A simple way to block gene flow is to have what was once a single population become physically separated into two populations. There are many ways this can take place: a fire or a landslide could break a habitat into two areas, a few members of a population could be dispersed to a new location far away from their founding population, and so on. Note that separation does not guarantee that speciation will occur – the two populations may reconnect soon enough that nascent genetic differences are again averaged out, or the two populations may track each other in parallel closely enough (either through chance, low genetic variability to start with, or due to similar selection pressures) such that no changes capable of blocking interbreeding arise before the populations come together again. What separation does achieve is the possibility of speciation. Now that the two populations are genetically isolated from each other, differences can begin to slowly build up. To return to our flip-book analogy, these two populations start from the same page, but then go their separate ways, with their average characteristics shifting slightly, but now independently from each other. Given enough time, the differences that build up may affect characteristics relevant to reproduction (i.e. genetic exchange): traits used for courtship, competition for mates, or behaviors suited for rearing young in a specific environment. These changes now begin to “genetically enforce” the separation of the two groups, even if they come into geographic contact again and some interbreeding does happen. Once these barriers are in place, the two groups no longer interbreed as successfully between groups as they do within their own group, and the groups are recognized as closely-related species. This process is called allopatric speciation (in ancient greek, allo = “different”, patrida = “fatherland”) indicating that physical separation was the initial barrier between members of the same population.

Separation by niche (sympatric speciation)

While allopatric speciation is relatively easy to envision, sometimes species arise without ever being geographically separated. If speciation occurs without separation, it is said to be a case of sympatric speciation (in ancient greek, sym = “same”). The availability of a new ecological niche can lead to sympatric speciation, as different subsets of a starting population exploit different environments within their geographic range. This process is not as abrupt as geographical separation, and it may take longer for substantial differences to arise that impede cross-breeding between the two sub-populations. Allopatric and sympatric speciation are in fact extremes on a continuum from no geographical overlap (allopatric) and complete geographical overlap (sympatric). In real life, a speciation event may lie somewhere in between.

Beyond the textbook

One real-world example of speciation that we have touched on before that exemplifies these complexities is Dolph Schluter’s work on the evolution of British Columbia’s freshwater sticklebacks: small lake fish descended from sea-dwelling ancestors. What we observe is lakes with stickleback species pairs, both descended from the same sea-dwelling ancestor. These same-lake species pairs were once thought to be classical examples of sympatric speciation, but further research has confirmed a more complicated picture, where two independent colonization events by the marine form occurred. The first colonization is a textbook example of allopatric speciation, where the marine form adapts to a new environment. The later, second colonization events introduce the marine form into lakes that already have the first species in a specific niche. The new invaders manage to persist in a different niche, one more similar to the one used by the marine form. Additionally, there is evidence of interbreeding between some species pairs, and in one lake the two forms have collapsed back into one species as a result of new environmental pressures. The lake forms are distinct from the ancestral marine form, and these species have been shaped by allopatric and sympatric factors. Some of the genes (and their various alleles) involved in the major differences between these species pairs have been identified. These genetic variations are present in the marine species, but in the lake species certain alleles have been nearly lost from the one species, and kept in the other, enforcing the physical and lifestyle differences between the two. These differences, in turn, raise a barrier to reproduction that largely keeps the species separate, though interbreeding remains possible, if unlikely.

Human speciation

The relevance of this discussion for human speciation may not be immediately obvious, but hopefully a few points are now more accessible. The first point is that speciation happens at the level of a population, as its average characteristics change over time relative to other populations from which it is reproductively isolated. There are other tools and techniques that geneticists can use to estimate how large a population was as it went through a speciation event, and we now know that our species has never been below around 10,000 individuals at any point in our evolutionary history. Some of those tools use genomes of our close relatives (e.g. chimpanzees) as a basis for comparison, while others look solely at genetic variation within our species, and thus do not require any “evolutionary” assumptions. These methods, though based on distinct methodologies with independent assumptions, nevertheless give the same answer for our population size over the last several million years. Darrel Falk and I have written about this evidence here on BioLogos in the past.

A second point is that this process proceeds through incremental steps. Like those pages in a flip book, each generation is nearly indistinguishable from the one preceding it, and characteristics change as averages, not as sudden shifts. Over time, these changes can build up and raise reproductive barriers between populations, producing new species. This suggests that many of the features we recognize as distinctly human arose incrementally. For some evangelicals, this understanding can be troubling if a more discrete event of “becoming human” was expected, or if they hold the view that natural explanations remove the need for God’s activity. One of the challenges for believers who accept the evidence for human evolution is mapping characteristics of theological importance, such as the declaration that humans are created in the image of God, onto this understanding of our biological history. This is an ongoing discussion, and one that has, and should, generate much discussion.

A third point is that as we became human, the population that eventually became Homo sapiens did not suddenly cease to interbreed with other groups. Like sticklebacks, our evolutionary path appears to be a mix of allopatric and sympatric speciation, with complex patterns of geographical migration and interbreeding with other hominid groups. We will have more to say about this in upcoming posts.


Dennis Venema is professor of biology at Trinity Western University in Langley, British Columbia. He holds a B.Sc. (with Honors) from the University of British Columbia (1996), and received his Ph.D. from the University of British Columbia in 2003. His research is focused on the genetics of pattern formation and signaling using the common fruit fly Drosophila melanogaster as a model organism. Dennis is a gifted thinker and writer on matters of science and faith, but also an award-winning biology teacher—he won the 2008 College Biology Teaching Award from the National Association of Biology Teachers. He and his family enjoy numerous outdoor activities that the Canadian Pacific coast region has to offer. Dennis writes regularly for the BioLogos Forum about the biological evidence for evolution.

Next post in series >


View the archived discussion of this post

This article is now closed for new comments. The archived comments are shown below.

Loading...
Page 2 of 2   « 1 2
Kirk Durston - #65365

October 3rd 2011

I am a little pressed for time today, as I am leaving tomorrow for a conference for the remainder of the week and urgently need to get a few things done before I leave. Thus, my comment below will be relatively short and be the only contribution I can make this week. I decided to focus on what I think is the more interesting of the two papers you link to above, the second one dealing with the supposed commonness of protein folding.

Biological globular proteins need to have stable 3D structure. To clarify, by ‘stable’, I mean that the structure needs to be repeatable in vivo, often to within a few angstroms in certain sections of the structure. It is highly probable that any random sequence will fold, but the probability of it folding into a stable, repeatable structure seems to be very low. The authors state that they did not perform any ‘direct assays for tertiary structure’, although they say that ‘it is difficult to imagine that the QLR proteins could self-assemble into stable oligomers in the absence of some stable tertiary interactions.’ As a result of my own work on sub-molecular relationships within the tertiary (3D) structure, the next largest structural units up from the individual amino acids are 3 to 4 residue clusters. These clusters are nested within higher order clusters to form short ‘branches’ which converge with other branches at ‘nodes’ to form sub-molecular modules which, in turn, fold together to form a structural domain. We have a paper in the final stages of formatting and submission showing how these sub-molecular relationships can be identified. As a result, the same sequence will give a repeatable secondary structure with high probability, which their QLR proteins do. Indeed, it would be surprising if they did not. The helices thus formed are likely going to fold into modules of repeatable mass, though it is less clear that they will have a repeatable structure, the number of variables are greater in a module than in a short length of secondary structure. The formation of stable oligomers is also not surprising. Their argument for a repeatable 3D structure would have carried more weight if they had attempted to crystallize at least one of the QLR proteins. Their QLR proteins do show evidence of having a molten core, which they discuss, which may explain a major difference between their protein and biological proteins which, in turn, underscores another key requirement of the type of repeatable structure required for life.

Let us grant for the sake of the argument, however, that at least one of their QLR proteins does form a stable, repeatable structure in vivo. I see a major problem with the conclusion that stable (repeatable) structures may be common in random sequence space. The problem is that they have radically simplified sequence space from 20 amino acids down to predominantly 3 and they have (intelligently) chosen those 3 to fall into the binary code strategy (in fact, they suspect their QLR proteins are forming the usual a4 four-helix bundle). They draw their sequences from a ‘random’ sequence space that has only three variables instead of twenty. There is a major problem with this; the more variables the more possible outcomes and the more difficult it is to achieve a repeatable fold. Their experiment simplifies folding by many orders of magnitude. It is similar to the difference between solving a set of equations with only three variables as opposed to solving a set of equations with twenty variables. As the number of variables are reduced from 19 or 20, the more unrealistic the experiment becomes insofar as addressing the real life problem of the origin of biological proteins. Finding stable repeatable folds in a 3-variable sequence space should be numerous orders of magnitude simpler than finding stable repeatable folds in a 20-variable sequence space. In other words, they have massively simplified the problem and should not, therefore, conclude that their experiment tells us anything about the ease of finding stable repeatable folds in 20-variable sequence space (and it appears that we need at least 19 of those variables for biological life).


Ashe - #65369

October 3rd 2011

It is not thought that all 20 amino acids appeared simultaneously on the scene, but that more limited libraries were available.



John - #65377

October 3rd 2011

Kirk wrote:
“Biological globular proteins need to have stable 3D structure.”

No. In fact, most don’t. 

“To clarify, by ‘stable’, I mean that the structure needs to be repeatable in vivo, often to within a few angstroms in certain sections of the structure.”

To clarify your ignorance of structural biology, take myosin. Its structure oscillates over time, so much so that it moves 100 Angstroms.

Or take collagen, which stretches.

“It is highly probable that any random sequence will fold, but the probability of it folding into a stable, repeatable structure seems to be very low.”

Biology is utterly dependent on the INstability of thousands of protein structures.

John - #65417

October 5th 2011

“John, you seem to have missed the significance of what I said by way of clarification.”

No, Kirk, it was easily recognizable as gibberish.

“I stressed the importance of repeatability. You may want to acquaint yourself with misfolded proteins, amyloidosis and prions, especially noting their effects, if this is something you are not familiar with.”

I’m quite familiar with them. They completely contradict your initial claim!

“As far as your response to the ‘few angstroms’, you may want to re-read what I wrote. Note the words, ‘often’ and ‘certain sections’. ‘Often’ does not mean ‘all’ or even ‘most’, and ‘certain sections’ does not mean the entire structure. Globular proteins usually need…”

But now you’re moving the goalposts. Your initial claim was simply, “Biological globular proteins need to have stable 3D structure,” with zero qualification. That claim is false. Biology depends upon instability in protein structure.

”… to fold into the same 3D structure (rather than misfiled)”

Proteins misfold all the time. Even when they are correctly folded, they do not have “the same 3D structure.” The vast majority have multiple 3D structures, Kirk. Please stop misinforming people about basic structural biology. How many different 3D structures of the globular head (motor) domain of myosins exist? How many are still being sought?

”… and certain sub-molecular components are extremely well conserved and need to exist within certain tolerances for key bonds to form. However, I fear you are missing the elephant in the room.”

Proteins generally have multiple structures. It’s how they work!

Kirk Durston - #65405

October 4th 2011

Ashe, I am prepared to grant that it is plausible to start off with fewer variables, but the problem is going from a 2 or 3 variable system to a 19 or 20 variable system. In general, simple things tend to de-stablize the more variables that are added to the system. We cannot take that transition seriously unless someone does the work to show it is even possible. The transition would involve horrendous problems even for a level of intelligence and knowledge considerably beyond where we are today, but to expect an unguided process to work its way through that transition before we even arrive at LUCA is a non-starter. As I said, before we can take that seriously, we need to do the science.

One indirect way to test the theory that the earliest proteins were built in a 3-variable sequence space is to take the universal proteins and show that for even just one, there is a functional subset that can be built with only 2 or 3 amino acids. That would be a start, but I really think we would need to be able to do that with all the universal ribosomal proteins, since they would be needed for replication. Having suggested this test, I have very little hope it would succeed on the basis of my work with several of the universal ribosomal proteins.

John, you seem to have missed the significance of what I said by way of clarification. I stressed the importance of repeatability. You may want to acquaint yourself with misfolded proteins, amyloidosis and prions, especially noting their effects, if this is something you are not familiar with. As far as your response to the ‘few angstroms’, you may want to re-read what I wrote. Note the words, ‘often’ and ‘certain sections’. ‘Often’ does not mean ‘all’ or even ‘most’, and ‘certain sections’ does not mean the entire structure. Globular proteins usually need to fold into the same 3D structure (rather than misfold) and certain sub-molecular components are extremely well conserved and need to exist within certain tolerances for key bonds to form. However, I fear you are missing the elephant in the room. To clarify, we have an enormous amount of sequence data now available to us. In the past, evolutionary biology has been able to survive on creative scenarios, suggestions, possibilities and a lot of ‘may haves’. This should not be confused with doing science. Now we have an enormous amount of sequence data that is, effectively, a record of hundreds of trillions of evolutionary searches randomly sampling and mapping the boundaries of folding, functional sequence space for protein families. We have enough data now to actually get a reasonable estimate of the target size of many protein families in sequence space. What I am saying is that the real science, based on real data, is effectively and systematically falsifying a fundamental prediction of unguided evolutionary theory ..... that an evolutionary search with an upper limit of 1043 trials, can find even the universal protein families .... forget about the rest. At the same time, we have powerful, positive evidence for an intelligent input to life ...... the high level of functional information encoded in the genomes of life. Effectively, we have the fingerprints of intelligence all over biological life. So there are really two elephants in the room, the ridiculous ineffectiveness of a ploddingly slow physico-chemical search engine to locate the folding, functional protein families in sequence space and the undeniable fingerprints of intelligence all over biological life in the form of functional information.


John - #65418

October 5th 2011

See above for the first part of my reply.


”…This should not be confused with doing science.”

Indeed. So why do all IDCreationists universally conflate producing rhetoric with doing science?

“Now we have an enormous amount of sequence data that is, effectively, a record of hundreds of trillions of evolutionary searches randomly sampling and mapping the boundaries of folding, functional sequence space for protein families.”

No, Kirk, because selection is involved. That makes your blanket use of the term “randomly” blatantly false.

“We have enough data now to actually get a reasonable estimate of the target size of many protein families in sequence space.”

Yes, and the data don’t support the notion of ID because very little of available sequence space has been explored. When real scientists explore it, it contains plenty of functional sequences.

“What I am saying is that the real science, based on real data, is effectively and systematically falsifying a fundamental prediction of unguided evolutionary theory …”

Straw man alert!

”… that an evolutionary search with an upper limit of 1043 trials, can find even the universal protein families .... forget about the rest. At the same time, we have powerful, positive evidence for an intelligent input to life ...... the high level of functional information encoded in the genomes of life. Effectively, we have the fingerprints of intelligence all over biological life.”

Yet those of us who actually do the science don’t see it the way that you do. In the end, all you have are empty assertions and a desperate need to misrepresent reality (most proteins do not have a single 3D structure) and the predictions of evolutionary theory.

Kirk Durston - #65422

October 5th 2011

I have appreciated the kind of questions and comments presented by Ashe. They are the kind of questions and comments that need to be raised and discussed. I am retiring from this thread, Ashe, but I have enjoyed and appreciated your questions and comments.

John, the tone and content of your posts suggests to me that you are in well over your head but are trying to make up for it with personal attacks and belligerence. Some of the things you have said demonstrate a real lack of knowledge about the subject you are portraying yourself as an expert in yet you seem absolutely convinced, in an Archie Bunker sort of way, that you know what you are talking about. I have found that ‘discussions’ with such people are completely unproductive..

If you wish to participate in a way that does not discredit yourself, I would recommend reading the posts that Ashe has contributed in this thread. A well-honed question or comment is much more effective that unprofessional hooting and chest-thumping.


beaglelady - #65424

October 5th 2011

Kirk,

On another thread Darrell Falk has confirmed that John is a researcher at a top institution.  Why not contact BioLogos to confirm this? Could it be that you can’t hold your own with an expert who knows the science?


John - #65429

October 6th 2011

Hi Beaglelady,

To give you an example from real people who do real science, the G proteins exist in two different structures, active and inactive. Their slow GTPase enzymatic activity is an extremely energetically wasteful way to make them act as metaphorical timers (the hydrolysis of GTP changes them from the active to the inactive structure) in signaling pathways.

The ubiquitous tools that cell biologists and pharmacologists use to study their functions are two mutants for each G protein—an activated one and an inactivated one.

These are valuablthe structures of the mutants are FAR MORE stable than the structure of the wild-type or normal protein. This is why I wrote that the function is utterly dependent on the instability of the structure. Kirk’s claim that “Biological globular proteins need to have stable 3D structure” is the rare exception. Normally, the polar opposite is true.

But we both know that reality is no barrier to Kirk’s rhetorical bluster!

beaglelady - #65438

October 6th 2011

Thanks for the explanation John. And thanks for keeping this blog honest! Please come by as often as you are able.


John - #65428

October 5th 2011

John, the tone and content of your posts suggests to me that you are in well over your head but are trying to make up for it with personal attacks and belligerence. Some of the things you have said demonstrate a real lack of knowledge about the subject you are portraying yourself as an expert in yet you seem absolutely convinced, in an Archie Bunker sort of way, that you know what you are talking about. I have found that ‘discussions’ with such people are completely unproductive..”

br>
Kirk, you are projecting. Your resort to unadulterated ad hominem is hilarious!

Page 2 of 2   « 1 2