Register | Sign In


Understanding through Discussion


EvC Forum active members: 65 (9162 total)
2 online now:
Newest Member: popoi
Post Volume: Total: 915,819 Year: 3,076/9,624 Month: 921/1,588 Week: 104/223 Day: 2/13 Hour: 0/0


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Author Topic:   Molecular Population Genetics and Diversity through Mutation
Modulous
Member
Posts: 7801
From: Manchester, UK
Joined: 05-01-2005


(1)
Message 5 of 455 (784833)
05-24-2016 7:41 AM
Reply to: Message 4 by Faith
05-24-2016 6:04 AM


Re: But I'm just as mathematically challenged as ever
. The selective or subtractive processes that bring about the new phenotypes toward a new subspecies HAVE to get rid of the genetic diversity that doesn’t support them.
This isn't quite true for a few reasons which reveal quite fundamental misunderstandings about what is happening.
So let's consider a foundation population on an island that is smaller than the parent population but all the alleles are represented in the foundation population. The environment is different on the island so the selective pressure is different. With the classic finches, there might be a pressure towards certain shapes of beaks, for example. So alleles which stand in the way of the trait we are looking at, reduce in frequency in the genepool. This is fine, it may result in the extinction of all of the involved alleles but harmful alleles can hang around.
On the whole though, let's agree there is some loss of alleles.
But while this is going on, and after the population beak size has stabilized, mutations are occurring. Other genes, less constrained than beak related ones mutate as normal, and the beak ones are still mutating during the selection process, so there were lots of alleles for medium sized beaks, now there are lots of alleles for thin sized beaks.
Let's say they grow to a reasonable size a new population splits off from them back to their homeland, where the original population lived, but is now extinct for convenience of this example. Medium sized are needed and there are no medium beak alleles!
Well it's not quite that simple, for a number of reasons but let's just suppose a small number of ideas that might give us reason to rethink the certainty of your position
a) The new population may survive with their less than optimal beaks for mutations to give rise to alleles that cause medium sized beaks. We know there is a mutational pathway that will allow it.
b) The new population may have broken copies of the medium beak allele which may allow for the re-emergence of the medium beak allele
c) The new population may, because of the mutations and variety building up back on the island and in the time before that when they were in the original parent population, have other strategies for survival open to them. Perhaps a shift to a heavier insect diet and away from crustaceans etc. While they are heading in that direction, maybe medium sized beaks won't even be advantageous anymore, even though the external environment is the same as with the parent population - the internal environment (ie the genotype) differs from the parent population.
But if selection eliminates genetic diversity you’d think that would be a caution right there. Additions CAN’T overcome this effect because subtraction is REQUIRED if evolution is occurring.
This presupposes that additions can't occur after subtraction when in fact they happen all the time. Each generation is an addition of mutational variety. Some varieties reproduce more than others. Some don't reproduce at all. Of those that do, their reproductive efforts are an addition of mutational variety, those that don't may represent a loss of variety. You can't just focus on where the water is leaking from and conclude it'll stop on its own soon enough if you don't note that there is more water entering the system from elsewhere.
There is no one selective pressure. You seem to be of the mind that 'evolution' starts with the culling out disadvantageous traits. There is no start point to evolution. We can draw arbitrary lines to discuss a narrative which helps us understand what is happening to varying degrees of depth, but evolution is happening all the time. Constantly. Every organism is a test, every birth a new trial run.
It's just interesting to us when the environment changes resulting in the population's phenotype changing. That environmental change can be changes in the environment or changes within the genepool (if allele X for gene 1 is lethal if in the presence of allele Y for gene 2, then how common allele Y is going to impact how common allele X can feasibly exist within the population. If something causes allele Y to become uncommon, then allele X may now increase in frequency without that impediment)
But evolution is about changes in the genepool, not the phenopool. The changes to the phenopool are caused by changes to the genepool (ie., an organisms physical traits are caused, in part, by the genes), but to understand evolution we need consider the genepool only. How advantageous or disadvantageous an allele is can then be abstracted from the question as to WHY it is advantageous (or not).
If you don't do this, you are right, it all looks a bit mysterious. But only looking at the general outcome is missing what's going on under the hood. Genotype and phenotype are loosely connected through an interaction with the environment. It's a chaotic system with chaotic outcomes.
You simply can't look at it with a non-mathematical eye and make reasonable predictions as you claim to do.

This message is a reply to:
 Message 4 by Faith, posted 05-24-2016 6:04 AM Faith has not replied

  
Modulous
Member
Posts: 7801
From: Manchester, UK
Joined: 05-01-2005


(1)
Message 7 of 455 (784843)
05-24-2016 11:59 AM
Reply to: Message 4 by Faith
05-24-2016 6:04 AM


Re: But I'm just as mathematically challenged as ever
Since maths is too much, maybe a graphical visualisation may help understand the same principles in a more intuitive fashion:
So the whole 3 minutes is interesting (the last is very interesting, but its less obvious why - sometimes less optimal solutions are what evolve), but the dynamic fitness landscape with three peaks is interesting as the species influence one another and seperate out into three groups on 3 peaks (aka speciation). So that's how it might work in simple terms, the question is about diversity.
In this 2d space, phenotypic diversity is a function of the area described by the creatures on the edge (that is to say, draw a line around all the creatures and calculate the area of the shape drawn). phenotypic diversity is a function of geneotypic diversity. So we can see that diversity may change depending on the landscape, but it can move from one peak to another with little drop in diversity anywhere.
In that final example, there wasn't much diversity. There was a very fit phenotype, but anything slightly different is worse (ie any mutation is a harmful mutation), but nearby there was a way to be slightly fitter than terrible and the population stabilized to the more diver solution as it was more stable and that's how the maths work out. They cluster to average fitness not optimal fitness, that way, mutations are less likely to be so regularly harmful.
Hardy-Weinberg
So let's consider a population that is static from generation to the next. Or a single generation if that's easier. Let's talk about beaks too for fun. 1 gene with two alleles. One allele is F and is dominant. Another is f and is recessive. The real world is more complex, but then maths has to come in so I'm trying to avoid that.
Because of the rules of Mendellian genetics, you can test the phenotypic frequencies and infer the genotypic frequencies. Unfortunately as I say, real life is more complex as traits don't typically have a single gene controlling them. But that's not too important.
In this static population there is nothing that affects allele frequency so once we have worked it out we're done. So what if something were to start affecting things, then what?
1. Natural selection
Natural selection will cause alleles that are 'fit' to increase in frequency and the less fit to decrease in frequency. If the reason the gene is 'fit' is because it helps to build a certain phenotypic trait, such as F and f, then we may see beak size frequency to also change.
It might turn out that if F is less fit, it goes extinct because it always gets expressed and so it is always 'exposed' to natural selection acting because of the phenotype. If f is less fit instead, it only gets 'exposed' when the organism has ff, so extinction is less certain. But it is possible that because of natural selection we lose F or f entirely.
This is represented by the peaks in the 2D diagram.
2. Sexual selection
Another selective force that can apply in some cases where mating is non-random, and allele that create sexy phenomes or phenomes that find other phenomes sexy may change the frequencies around. Again an allele could be lost.
3. Drift
Can only act to reduce allele frequencies to extinction as described in OP
4. Gene Flow
Alleles sporadically popping into or out of the population, such as when one population becomes physically isolated or when a previously isolated group's isolation ends such as with immigration.
For a given population this can add or remove alleles.
5. Mutation
Everytime a mutation happens to an allele it creates a new allele unless it happens to mutate the allele in such a way so as to be the same as an already existant allele. How often an allele mutates varies based on a number of factors but reasonable estimates can be made based on observation and mathematics.
Each offspring in the 2d landscape video is slightly different from its parent. This allows for the population as a whole to take random walks towards stable peaks in the landscape. Without the mutation there is no searching through possible trait combination and no adaptation.
In your version of evolution a new peak emerges and if it somewhere near where the population is in this abstract space, everything not on the peak is killed off and lost forever. With mutation, we can see it isn't so obvious from a casual inspection of life.
The Fitness landscape
The 3 minute video just looks at mutation and selection of two traits, which allows for a 2d shape or 'landscape' to created visually for what combination of those traits is most 'fit' and to see that because new variations are mutating, the population can move from peak to peak in certain circumstances. It's a random search through a 2d space. But if we're doing this by traits, then there are a lot more than two traits, so the reality is that we're looking at a several-thousand-dimensional landscape, which shifts and change dependent on what's going on within it. As I said, it's complex and chaotic, but it's a chaotic system searching (through logical necessity) for some kind of optimal stable equilibrium and if the circumstances stop changing life might find itself in equilibrium for a long time, before getting knocked out of it again by a changing landscape.
The question you raise is, is there enough mutation to overcome the selection process? Well it's a good question, and it's not a trivial one too. All I can tell you from here is that it is a mathematical and empirical question. The pros tell me it can work and here's how. You tell me it can't because it can't. I'm not a scientist, but those guys seem to have better reasons for saying what they do than you do.
I see no reason a population cannot bounce between two peaks forever as long as the peak changes are slow enough compared to the mutation rate to allow for 'tracking the peak'. You think they'll run out, but slow the selection pressure changes and they just keep getting more and more. So unless someone has done the maths that says otherwise, you have no basis to intuit that it must result in long term depletion of diversity.
I will grant you this though, most species are extinct. Some extinctions had survivors in the form of daughter species, many did not. Those that left nothing downstream were not adapting to the changes in their fitness peaks fast enough. They lacked a pathway through the space to get to a new peak before all of their genes were selected out in total loss of diversity mode.

This message is a reply to:
 Message 4 by Faith, posted 05-24-2016 6:04 AM Faith has replied

Replies to this message:
 Message 8 by Faith, posted 05-24-2016 4:29 PM Modulous has replied

  
Modulous
Member
Posts: 7801
From: Manchester, UK
Joined: 05-01-2005


(1)
Message 13 of 455 (784852)
05-24-2016 5:14 PM
Reply to: Message 8 by Faith
05-24-2016 4:29 PM


Re: Fitness graphic
HBD posted that graphic quite a while ago and it's just as mystifying to me now as it was then, as mystifying as math.
It measures two traits that result in either faster speed or larger size. The further North a creature is in this place the faster it is. The further east it is the larger it is. A creature in the North East corner is very fast and very large. In the South West corner they are slow and small.
It's just a graph. But each point on the graph has a 'fitness' This is represented as height to create a landscape of peaks and valleys. Peaks are high fitness areas of the graph. Valleys are low fitness areas.
. There are conditions in which populations are stable, that is, NOT EVOLVING, which is what Hardy-Weinberg is describing.
No such conditions exist in reality. Instead the population reaches an evolutionary stable state. Drift and mutation are still happening and selection ensures any mutations that push organisms too far away from the peak will decrease in frequency relative to those closer.
It's not Hardy-Weinberg.
It is actually described AS not evolving.
Maybe, but its still part of evolution whether the phenotypes are changing or not.
Again, what I am focusing on is what happens when you are getting the active production of new phenotypes in reproductive isolation, because that is the clearest expression of evolution, usually believed to be THE way all life evolved from former life.
And this is shown in the video. It shows a species at minimal fitness with three different 'niches' or peaks in fitness open to them, they split into 3 groups and the variety of traits is not impacted for very long.
But the fact is that this leads ultimately to genetic depletion, yes even with all the additions you can throw at it -- as long as new phenotypes are being produced you are getting a loss of genetic diversity.
Addition followed by subtraction followed by addition....
I'm telling you that people that know the maths say you are wrong, you can't understand the maths.
I give you a video that explains the concept in terms of two traits using a visualisation rather than raw maths you say
What that graphic is designed to show I really have no idea, however.
It shows how a population's traits can change without losing genetic variabiality. Ie., it shows your intuitions on this subject are not to be trusted as it is as a fact, possible for a populations traits to evolve without losing genetic variability.
Whatever else you say, some of the evidence that you always beg for is being presented. As ever you claim not to understand it. But at the same time you claim to know better, without providing any reason.
I can agree that natural selection is one way genetic diversity is reduced as new phenotypes are expressed, making it one version of what I'm arguing. I just think it's clearer to emphasize the factor of reproductive isolation of a subpopulation, which is also brought about by natural selection but can be seen a lot more clearly in the example of geographic isolation.
That's fine, but it's important not to lose sight of the big picture by focussing only on a part of it. The visualisation in the video can be seen in terms of this. Look at the dynamic 3 peak example, the west most peak is the fittest way to be while living in the home area and the other two peaks represent other, isolated areas. That way the parent population is seen to evolve towards fitness, and the two daughter species evolve towards the other peaks - simulating the example you suggest. And as the visualisation of the mathematics shows, diversity is not lost intrinsically lost.
Look at the last example, not only is diversity not lost, it's gained. Diversity is represented by the area of the population in the graph.

This message is a reply to:
 Message 8 by Faith, posted 05-24-2016 4:29 PM Faith has replied

Replies to this message:
 Message 14 by Faith, posted 05-24-2016 5:23 PM Modulous has replied

  
Modulous
Member
Posts: 7801
From: Manchester, UK
Joined: 05-01-2005


Message 16 of 455 (784855)
05-24-2016 6:28 PM
Reply to: Message 14 by Faith
05-24-2016 5:23 PM


Re: Fitness graphic
That's how increases in genetic diversity are proved by math, by simply assuming it's possible when it's not.
Nope. The mathematics does not make this assumption. The maths tells us under what conditions of selection and mutation an increase in diversity is possible, indeed inevitable. It also tells us under what conditions genetic diversity will decrease.
The graphic in your post is no doubt a case of the same error.
I'm going to say that your errors are bigger than its errors are.
Oh yeah yeab yeah, EVERYTHING is evolution, yeah yeah, but that's why I keep saying ACTIVE evolution, because I'm describing the specific process where you are getting new characteristic phenotypes and ultimately a new subspecies based on those phenotypes.
Come up with all the new terms you like, the fact is that the net result of the process that evolutionists say is responsible for biological change can be an increase in diversity. Maybe some part of it is subtractive, but some parts of it are additive.
If you want to argue that the mathematics that supports evolution is wrong you can't just say 'I assume the mathematics behind the evolutionary process biologists are talking about proves it is wrong'.
Until you can watch that video and understand what it's saying and how it relates to your objection I can't present the argument any better. I am happy to answer questions, but simple denial makes for a boring discussion. I didn't understand this stuff when I started at EvC Forum, you can learn it if you are motivated, but I feel your religious objections may interfere with that motivation.


Here is a diagram
This is a simplified version of the graph in the video. Let's, for the sake of argument, say it measures 'beak thickness'. The horizontal axis represents thickness, the y axis represents fitness. If the population is at the point where the red arrows start, it will tend to go towards peak B.
Imagine every single member of the population marked on the graph. Some will have thicker beaks some smaller. The phenotypic variation is measured approximately by the length measured from the thinnest beak to the thickest beak. The ones close to B reproduce more and so the population 'crawls' towards B over time.
If a subpopulation breaks off and goes to say, an island, the curve of the graph may well be different. Let's make it simple though, just for illustration purposes. Let's say the curve is basically the same, except that everything to the left of B is 'pushed up' the y axis relative to B (so B can drop it doesn't matter exactly). Now the population that split off may be less diverse, but through mutation and selection it too will crawl up towards A instead of B. The length of the population can remain the same, it's just evolving in a different direction because the fitness landscape differs on the island. The length can remain the same under certain conditions, this is mathematically proven. Now the two populations are separate and Active evolution has occurred, but the length/width of the population on the graph (distance between thinnest and thickest beak thickness) can, under certain circumstances stay the same in both populations. Thus it is possible for diversity to remain. It's just a diverse range of thinner beaks rather than a diverse range if middling beaks.
The only real challenge you face is now to show that those certain circumstances can't exist. We've looked and as far as we can see, they can.

This message is a reply to:
 Message 14 by Faith, posted 05-24-2016 5:23 PM Faith has not replied

  
Modulous
Member
Posts: 7801
From: Manchester, UK
Joined: 05-01-2005


Message 34 of 455 (784892)
05-25-2016 8:03 AM
Reply to: Message 27 by Faith
05-25-2016 12:18 AM


Re: But I'm just as mathematically challenged as ever
You will get new phenotypes simply from a new set of gene frequencies. Not new in the sense of novel but new to the population. If the parent population was characterized by brown fur the new subpopulation may have more alleles for gray fur and after some period of mixing of the new gene frequencies over some number of generations the new population will now look gray or mostly gray, differing from the original population by that particular trait.
Which is what would happen if mutations did not happen.
Let's look at a simple example population with 2 members (sexually reproducing). In this population there can be NO MORE than 4 alleles in the entire population for any gene. Without mutation there can never be any more alleles in this population. There could be less, but there cannot be more.
However an error occurs when copying the DNA during reproduction, and it mutates the gene we are looking at. Now there are 5 alleles.
In a Hardy-Weinberg equilibrium (no selections, no drift etc) with only mutations happening the number of alleles very very probably goes up each generation. Agreed?
In an analogy, this would be like photocopying something. Each photocopy is going to be slightly different than the last as noise in the machine and the optics results in speckly dots starting to appear in the copies. There's no selection, and once a dot is in place it stays (no drift), so each copy is different, there is much variety. Right?
But to the general point, the whole genome should have new gene frequencies as a result of the population split...all the alleles throughout the genome should have undergone the same changes: some higher frequency than they were in the original population, some lower, including many also not changed or much changed in frequency.
It may be a terminology confusion of your own, or it might be indicative of something more. A genome does not have gene frequencies.
A population split does not imply new gene frequencies NECESSARILY although, due to the number of genes and the law of large numbers they are PROBABLY going to vary slightly, depending on actual numbers. If the subpopulation is a perfectly representative sample, there will be, by definition, no change in allele frequencies, this is not likely - but the subpopulation will be a sampling of the main population so the frequencies are likely to be similar - especially for the main alleles. In order to not die off it needs to be large enough to be sustainable, and so we can discount small populations from consideration.
So a subpopulation of a predominantly brown fur allele population is likely to mostly have brown fur alleles too.
Think of it in a different context. Let's say I were to pick 1,000 Americans at random. If I were to ask you, "How many of them will answer, "Yes", to the question, "Are you a Christian?"", what would you say? Well the real number isn't important here, let's say 70% Why? Because just about every random sample we take gives us approximately a 70% "Yes" response. We know the sample is a reasonable approximation to the whole.
Obviously geography plays a role in both cases, and this can lead to some differences - but the heavily represented alleles are likely to make it across and still be one of the most significant alleles in the subpopulation. The sampling error we get through just picking a population from a geographic area is more likely to affect alleles that aren't common to begin with rather than ones that are.
But as to mutations: in every discussion I've ever read, mutations are treated as sort of an article of faith, their actuality not being demonstrated, or the mere presence of a newly expressed phenotype gets it called a mutation without warrant. It’s science that describes them as so predominantly either deleterious or neutral and so extremely rarely of any value to the organism, but the obvious conclusion isn’t drawn from those facts.
Actually the obvious conclusion IS drawn from these facts. Equilibrium is the norm.
Not only is a beneficial mutation rare
The problem is that the rate of beneficial mutation is not fixed. At times of ACTIVE evolution the number of mutations that would be beneficial goes up. Some mutations that were harmful before ACTIVE evolution began are now considered beneficial because of whatever started ACTIVE evolution (eg., a new location with different challenges). This allows the populations to 'crawl' towards a new local and optimal set of phenotypes with the same variety.
This is shown visually on the video I posted earlier. I again urging taking a look at it to get a grasp on the principles. Once you get those, then we can look at the evidence to see if that's how it works in nature. When the populations are sitting on the peaks of the mountains, any mutation that would knock their kids of their peak is considered harmful. The steeper the sides, the more harmful mutations exist compared to neutral mutations. By definition, there are few beneficial mutations, and for an organism right at a peak fitness - there are zero beneficial mutations. All mutations are necessarily worse that the best possible.
ACTIVE evolution occurs when the definition of 'best possible' changes by a significant degree for enough generations. During ACTIVE evolution some organisms have traits that are closer to the 'best possible' than others, so these do better and reproduce more. The population then 'clusters' around the 'close enough' group in the next generation as the close enoughs have a naturally imposed bias towards reproducing offspring more so there will be necessarily more of them 'close enough' in the next generation.
The next group also is clustered around here, they may hover around here, but the tendency for a while is to weed out all the old optimal alleles and we have a population that is a little phenotypically different, but growing in diversity. New combinations of existing alleles will account for some of the variety, of course, but some alleles are getting lost and whenever an offspring has a mutation in the regions in question it pushes it closer to 'best possible', and that biases the next generation towards being a bit closer on the whole to 'best possible', and this intrinsic bias means there could be a 'mutational pathway' (a chain of one mutation to the next) of beneficial or neutral mutations that exist between the old optimal and the new optimal and if there is, by randomly trying all the possibilities - life will either find that pathway or get as close as it can.
the fact, that I didn't make up, that most are by far deleterious or neutral and anything at all beneficial is rare to the point of vanishing, means to me that mutations contribute nothing to the evolution of new varieties.
This is true only in equilibrium. Not during ACTIVE evolution. That's the key takeaway from my post.
But evolution doesn't continue once a population is established from its particular set of gene frequencies after some number of generations of inbreeding. It could even settle down to a Hardy-Weinberg stability for some great period of time.
Again, Hardy-Weinberg isn't and cannot be real. It would imply the laws of thermodynamics are false. It means the genome never gets a harmful mutation, a neutral mutation or a beneficial one. It means that the death of an animal is never related to its genetics. It's just not going to happen.
What you mean is an Evolutionarily Stable State
quote:
"A population is said to be in an evolutionarily stable state if its genetic composition is restored by selection after a disturbance, provided the disturbance is not too large. Such a population can be genetically monomorphic or polymorphic." --Maynard Smith (1982).
Again, just a minor quibble, but the possibility for misunderstandings for technical terms can result in unnecessary bickering, I find. I'll try to remember if you are referring to Hardy Weinberg you probably mean ESS instead. During ESS, most mutations are harmful (hence 'restored by selection after a disturbance') or neutral. This is what we see most of the time for most traits.
Watch the video Faith, seriously. It's the best explanation it's only 3 minutes long. I've watched hour long documentaries when you entered them into discussion as sources, if you are unable to watch youtube videos such as https://www.youtube.com/watch?v=4pdiAneMMhU are you able to watch videos in any format. I'm very keen because even if you disagree with evolution, understanding what's happening in this video will allow us to have mathematical discussions using English so much easier.
If the video is still nonsense to you, here is another one that takes more time to explain the concept, and it doesn't use biological evolution as its only example.

https://www.youtube.com/watch?v=JPTIZbGGYmk
Edited by Modulous, : No reason given.

This message is a reply to:
 Message 27 by Faith, posted 05-25-2016 12:18 AM Faith has not replied

  
Modulous
Member
Posts: 7801
From: Manchester, UK
Joined: 05-01-2005


Message 57 of 455 (785038)
05-27-2016 12:41 PM
Reply to: Message 54 by Faith
05-27-2016 4:31 AM


Re: You are looking at the wrong part of the system
Not about how new gene frequencies from a population split bring about phenotypic variation.
Gene frequency changes from a population split don't result in phenotypic variation. If variability is lower, it simply restricts the possible phenotypes that can be made and it may make some fairly different looking phenotypes more probability of being expressed through recombination, but this isn't the full picture.
For reasons that can be explained using Game Theory, a mixture of phenotypes is likely to reach a stable equilibrium in a population. There are some great examples of these 'mixed phenotypic strategies' in the world of fish and kleptogamy. The idea is that when males typically spend some energy growing large and brightly coloured and doing flashy things to get clearance from a female to fertilise her eggs, there is an alternative strategy available - to not bother with all that - stay small and female looking sneak in and fertilise the eggs while the other fish are ritualising. Of course, if EVERYBODY did that, the females would be able to fend them off, and males would be competing again. But if some people are doing it, a balance can emerge where the strategy works if say 10% of fish are sneaky, but if the next generation has 15% sneaky fish, they all do worse for it, reducing their numbers.
This isn't evolutionary by the way, sneaky fish have the same genes as straight players, the phenotypic changes but it shows in a very stark way that the success of certain phenotypes can influence the success of another within a species which results in a balance of strategies where presumably both strategies work out as being approximately equally successful.
Let's say brown fur/grey fur is in this category and it's influenced by a single gene F. F makes brown and is dominant. f makes grey and is recessive. Perhaps brown fur is good for hunting wheras grey fur is bad for hunting but is 'sexy' to female foxes because if you can live to reproductive age AND have grey fur, your other genes are probably pretty good to compensate.
We have a stable population that is not experiencing phenotypic change or genotypic change. The population has 100,000 members and we find there are 9,000 grey wolves when we count them. Using the maths of Hardy Weinberg we calculate that this means the Allele Frequency is 70% Brown Fur and 30% Grey Fur.
We watch, and occasionally there are 10,000 grey wolves, then 8,000 but it bounces around the 9,000 point. The sexiness of the gene makes sure that 30% of wolves have it, but when it gets too common the disadvantages to hunting become a bigger factor and this pushes them back to lower numbers where sexiness matters more which pushes them up and so on.
Even if a subpopulation of say, 10,000 wolves of 3,000 brown and 7,000 grey were to somehow split off, everything else being the same, if this population grows to 100,000 we'd expect there to only be 9,000 grey wolves. Brown wolves do very well at hunting in a grey wolf world and while they get more of the girls when they grow up, more of them die trying to grow up than brown wolves.
So evolution doesn't happen by pushing the population out of equilibrium. The population just tends towards the same equilibrium. This is evolution, but it's the conservative part of evolution, tending towards the current equilibrium.
Forget the subpopulation thing now, and we'll watch how evolution is said to happen.
The equilibrium point shifts. Right now, the balance is between two opposing forces: Hunting prowess versus sexual attraction. So let's flip things around. Let's say the forest land dies away and the area the population lives in becomes rockier and more prone to snow. Climate change it seems. Grey fur now aids in hunting, but brown fur is considered sexy because of the handicap effect. The dominant/recessive stays.
So the next generation is going to have more f alleles and F is going to decrease. f alleles are getting the big benefits of surviving to adulthood, and although F loses that, it gains something by being sexy. The next generation has 40% f rather than 30%. What does the population look like? Now there are 16,000 homozygous f and thus Grey Wolves rather than 9,000.
Selection carries on, pushing the population towards the new equilibrium. Grey has more force than Brown in the struggle and pushes it back....
When 50% of wolves carry f there are 25,000 grey wolves in our 100,000 population
When 60% of wolves carry f 36,000 grey wolves
At 70% we reach approximately 50,000 grey 50,000 brown
At 90% we reach 80,000 grey wolves
When 96% of the population has f, then we see approximately 92,000 grey wolves and 8,000 brown wolves and the phenotypic balance of sexiness and hunting skill is restored, the allele frequencies have change from 30% f to 96% f but we still have both F and f. Homozygosity has increased (and thus variability has decreased). But then, if the process was reversed Homozygosity would decrease.
So randomly 'selecting' a population with different frequencies out of a larger population doesn't cause evolution, isn't said to be the main element of evolution. Changes in the equilibrium point, results in selection and this can result in either a more homozygous or less homozygous population around that loci.
Your list was of ALL species within a species or I guess family. You listed ALL the different breeds of dogs and claimed no loss of genetic diversity. That would only be true of the entire family of dogs, because separate breeds do in fact have sharply reduced genetic diversity and where selection has been severe genetic depletion that has brought about genetic diseases.
quote:
According to the pairwise genetic distances, Greyhounds and German Shepherds had longer diverse evolutionary histories than Greyhounds and Labradors or Labradors and German Shepherds. Although a few breed-specific alleles were observed, the significant differences between breeds are in their relative frequencies and distribution of the alleles across a locus. None of the three pure dog breeds corresponds to Hardy-Weinberg equilibrium. A considerable reduction in intrapopulation variation was observed within three pure breeds, compared with the population of individuals belonging to 15 dog breeds. This reduction was especially pronounced in the Greyhound breed, which expressed the lowest degree of variation. Intrapopulation variations of Labradors and German Shepherds did not differ significantly, that of Labradors being only slightly higher. The intra-species variation of dogs is lower than in humans, mouse, or rat, but similar to that in domestic animals, probably reflecting similarly high inbreeding coefficients.
Domestic animals have other effects on variability, especially when you consider the Platonic fantasy that the 'purebreed' culture revolves around. Ultimately the differences aren't because alleles are lost between breeds. Some alleles do get lost due to stringent breeding programs which are essentially replications of the Noachic flood bottleneck.
For example:
quote:
All modern Thoroughbreds can trace their pedigrees to three stallions originally imported into England in the 17th century and 18th century,
Thoroughbred - Wikipedia
quote:
During the 1880s, the 3rd Earl of Malmesbury, the 6th Duke of Buccleuch and the 12th Earl of Home collaborated to develop and establish the modern Labrador breed.
Labrador Retriever - Wikipedia
Most labradors descending from an animal we have a photograph of:
Extrapolating back from the loss of genetic diversity brought about by selection, random or artificial or natural (it's all the same effect), I add back the genetic diversity lost down the generations and arrive at LOTS of genetic diversity at the starting point from which all the types and breeds descended. A lot more heterozygosity I've many times suggested.
The problem is that the most heterozygosity there can be in a population is 50%. The maximum number of alleles in a population is determined by its size - there can only be at most two alleles per individual. Take any population today and go back in time and the population collapses back towards some smaller pool of common ancestors (for you, this is the Flood for the sake of ease). In this scenario everyone may well be heterozygotic, but there are only two alleles per individual. So maximum 12 alleles in the Noachic ancestors. There's a lot of variety there, but I don't think it's enough to account for what we see. Particularly that there are more than 12 alleles for some genes.
So the problem is that your model ends up with there being less alleles in the past with no way for alleles to be added...
WHAAAAAAT? WHAT ARE YOU TALKING ABOUT? "Shuffling alleles around" proves you have no idea what I'm talking about. I've explained my idea about how we got from the ark to today many many times, and explained it again above.
Indeed, the problem is that you need to introduce variability into the system one way or another to explain what we see.
You have *something God related no doubt*
We have observed random noise.
But random noise is all we need, because there are those equilibria and gradiants towards them and a tip in the wrong direction does nothing, but tip the population towards the gradiant and it settles at that new equilibria.
Both explain what we see, and you can't just dismiss the noise.
It would be if you understood what I'm talking about and looked in the right place for the right evidence. It would also help if you understood that I'm talking about a TREND in that direction.
Yes, but the TREND of genetic variability depends on a number of factors and you are only looking a part of it.
It's like looking at a TV and saying that the image on the screen is about vanish because there is a way for the electricity causing them to escape the TV without considering that its part of a circuit and more electricity is entering it.
Edited by Modulous, : No reason given.

This message is a reply to:
 Message 54 by Faith, posted 05-27-2016 4:31 AM Faith has replied

Replies to this message:
 Message 59 by Faith, posted 05-27-2016 1:02 PM Modulous has replied

  
Modulous
Member
Posts: 7801
From: Manchester, UK
Joined: 05-01-2005


Message 68 of 455 (785072)
05-27-2016 2:20 PM
Reply to: Message 59 by Faith
05-27-2016 1:02 PM


Re: genetic diversity on the ark
You act as if I'd never discussed this.
Not at all, I'm just pointing out that from an allele standpoint you have the issue of there being fewer as you go back so you still need some way for them to have got here one way or another
I do need to account for the extra alleles that now exist for a single locus nevertheless, so I've considered the possibility that there is some kind of mutation, just not the random accidental kind.
Indeed. Or as I said
quote:
Indeed, the problem is that you need to introduce variability into the system one way or another to explain what we see.
You have *something God related no doubt*
We have observed random noise.
not the random accidental kind.
But why not? At least for microevolutionary events. As a real example take this:
quote:
The body hairs on the dorsal surface of Salukis and Afghan Hounds have both phaeomelanin and eumelanin portions, even though they had an a(t)/a(t) genotype at ASIP. In addition, all had at least one copy of a newly identified mutation in MC1R, g.233G>T, resulting in p.Gly78Val. This new allele, that we suggest be designated as E(G), is dominant to the E and e (p.Arg306ter) alleles at MC1R but recessive to the E(M) (p.Met264Val) allele.
A new mutation in MC1R explains a coat color phenotype in 2 "old" breeds: Saluki and Afghan hound - PubMed
In English, they note a gene that affects coat colour have a point mutation from a G to a T. You can see this further here:
UniProt
10         20         30         40         50
MNIFRLLLAT LLVSLCFLTA YSHLAEEKPK DDRSLRSNSS VNLLDFPSVS
        60         70         80         90        100
IVALNKKSKK ISRKEAEKKR SSKKKASMKN VARPRPPPPT PCVATRNSCK
       110        120        130
SPAPACCDPC ASCQCRFFRS ACTCRVLSPR C            
a mutation that turns this into
10         20         30         40         50
MNIFRLLLAT LLVSLCFLTA YSHLAEEKPK DDRSLRSNSS VNLLDFPSVS
        60         70         80         90        100
IVALNKKSKK ISRKEAEKKR SSKKKASMKN VARPRPPPPT PCVAT[color=red]C[/color]NSCK
       110        120        130
SPAPACCDPC ASCQCRFFRS ACTCRVLSPR C       
Arginine -> Cysteine will make an Alsatian/German Shephard black.
Interesting if maximum heterozygosity is 50% as you say.
Basically, yes.
Imagine there are two heterozygotic individuals, call them Adam and Eve. OK, I've proven the thesis wrong as there is 100% heterozygosity. But they mate what are their babies?
GG
gg
gG
Gg
50% of them will be heterozygotic. Obviously chance plays a role, but the trend is biased towards 50% heterozygotic as a maximum.

This message is a reply to:
 Message 59 by Faith, posted 05-27-2016 1:02 PM Faith has not replied

  
Modulous
Member
Posts: 7801
From: Manchester, UK
Joined: 05-01-2005


Message 70 of 455 (785078)
05-27-2016 3:01 PM
Reply to: Message 65 by Faith
05-27-2016 1:48 PM


Re: You are looking at the wrong part of the system
ABE: You aren't going to get "new" genotypes, just a new frequency of genotypes. Where you had, say, 90% bb, 8% Bb and 2% BB in the original population, you could have 20% bb, 70%Bb and 10% BB. Parent lots of blue eyes, daughter an increase in brown eyes which should increase even more down the generations.
This isn't how it works. The situation of '90% bb, 8% Bb and 2% BB' is not in Hardy Equilibrium. If this population has children, they will have different frequencies, which is why it is not equilibrium. Let's say 90% of the population has blue eyes. To get this the 'b' allele has to be 95% of all alleles. About 9% of the population will have Brown Heterozygous and 1% will be homozygous. This is the only way for blue eyes to be in Hardy Weinberg Equilibrium and be expressed in 90% of the population. So the real Hardy equilibrium is either 90% blue 10% brown OR more accurately 90%bb 9%Bb 1%BB
You pick a subgroup which has 20% blue eyes and 70% brown heterozygous. This isn't Hardy Weinberg so it tends towards this equilibrium. Let's say it settles down to 16% blue eyes and 48% brown eye heterozygous. This is a new equilibrium and instead of b having a frequency of 95%, b now has a frequency of 40%.
That's Hardy-Weinberg. The gene frequencies change between these two states, but that is because of emmigration, something which can interfere with the equilibrium and yes, emmigration is ONE of the things which causes frequency change but it is not the only one, and it its not a big one because other than random chance the only way for a large group to have disproportionally blue eyed people is if there is some naturally occurring selective force.
Adding to Hardy Weinberg with only natural selection, sexual selection drift and emmigration will result in subgroups having certain alleles unavailable to them, resulting in a tendency towards nothing.
You can have 1% blue eyes, 90% blue eyes and anything else in Hardy Weinberg equilibrium. Darwin's contribution was essentially to say (though he didn't know it at the time) that although any Hardy Weinberg equilibria is possible in nature, nature prefers some equilibria over others. Sometimes nature's preferred equilibria is 100% 0%, sometimes it is something else. This 'preference' is natural selection.

This message is a reply to:
 Message 65 by Faith, posted 05-27-2016 1:48 PM Faith has replied

Replies to this message:
 Message 71 by Faith, posted 05-28-2016 12:03 AM Modulous has replied

  
Modulous
Member
Posts: 7801
From: Manchester, UK
Joined: 05-01-2005


(1)
Message 79 of 455 (785146)
05-28-2016 10:01 AM
Reply to: Message 71 by Faith
05-28-2016 12:03 AM


Re: You are looking at the wrong part of the system
Dear dear Moddy, I said nothing about H-W equilibrium.
quote:
When you have a population of over a million individuals, say of black wildebeests, you are probably getting something like Hardy-Weinberg equilibrium in reality.
quote:
This quote from the Wikipedia article on Genotype Frequency doesn't treat the H-W equilibrium as an ideal but as a reality.
If for some reason it SHOULD be in equilibrium, though I don't know why it should, then adjusting the percentages is fine with me.
The reason I posted that was not to be pedantic, per se, but to try and inspire you to think about this very point.
Are the gene frequencies changing from generation to generation in the parent population in the scenario you envisage? If the answer is basically 'no' then they are at equilibrium. The default equilibrium for a population to be in is the H-W equilibrium. This is because of the maths involved.
You can play with the maths at this website:
Hardy-Weinberg Equilibrium Calculator | Science Primer
You enter the allele frequencies and it tells you how those frequencies can exist in a population that is not evolving. That is, it will tell you the distribution of heterozygotes and homozygotes.
This is the H-W equilibrium when the allele frequencies are 50/50
p is one allele q is the other. p squared means it has two p's and therefore is homozygous. 2pq covers the heterozygous cases. Here we see that if Blue Eyes gene is 'p' then if it has 50% frequency in the population, it will be homozygous and expressed only 25% of the time. When the distribution is even the dominant trait gets expressed 75% of the time. It's a little like the maths when you flip two coins. 25% of the time you will get two Heads, the rest will have at least one Tail.
Anyway, from the parent population you propose a split off:
quote:
20% bb, 70%Bb and 10% BB
Which is not a Hardy-Weinberg Equilibrium. So because of Mendel's laws of inheritance (totally evolution neutral stuff, no shenanigans) the next generation will be different. The exact details depend on information about our population we haven't defined, but I will illustrate the point thusly.
First of all, there's too much heterozygosity, that's definitely going to change next generation. Suppose it drops to 40%Bb leaving us with 35%bb and 25%BB (I distributed the 30% evenly just for ease). This isn't quite stable so it changes to 55%Bb 25%bb 15%BB. Still not stable the next generation comes out to
Which is H-W equilibrium of 60% blue eyed genes, and this is where it will stay until something changes. That's just how Mendel's work applies to large populations. This is a gene frequency change. It is by migration alone. It is not a sufficient picture of evolution.
Parent lots of blue eyes, daughter an increase in brown eyes which should increase even more down the generations.
And that's the point of equilibria, there is no 'should increase even more down the generations' unless something makes that happen. The laws of inheritance provide a default frequency distribution, there is no tendency towards loss of an allele through this process.
from which the daughter population randomly selected most of the B's which would give it a new set of genotypes
But how would that happen? And why couldn't that daughter population later have a split which 'randomly selects' most of the b's?
There's no tendency to loss of allele here.
The only time allele loss happens is when the migratory population happens to have 0 of an allele represented. This can only realistically happen if the allele was not distributed evenly throughout the population. An allele can only realistically not be distributed evenly through the entire population if it is an allele that is new to the population.
So again, how does a daughter population migrate with such radically different allele distributions? Why can't a granddaughter population use this same mechanism to restore the frequencies? Where is the allelic loss in this situation? I think you are picturing domestic breeding too closely, where daughter populations are chosen on purpose towards some end.
HBD's point is that the genotypes in your parent population are:
BB
bb
Bb
bB
And the genotypes in your daughter population are the same. How the genotypes are distributed in the population is different, but the genotypes are the same.
Edited by Modulous, : No reason given.

This message is a reply to:
 Message 71 by Faith, posted 05-28-2016 12:03 AM Faith has not replied

Replies to this message:
 Message 80 by NoNukes, posted 05-28-2016 3:23 PM Modulous has replied

  
Modulous
Member
Posts: 7801
From: Manchester, UK
Joined: 05-01-2005


(1)
Message 96 of 455 (785214)
05-30-2016 9:41 AM
Reply to: Message 80 by NoNukes
05-28-2016 3:23 PM


Re: You are looking at the wrong part of the system
I'm a bit suspicious of this idea. Sometimes populations do not intermingle thoroughly even over large periods of time.
This is just a matter of the 'largeness' of the time, and definitions of 'population'. Subpopulations that don't share genetic material with other parts of the population might simply not be considered part of the population. This would be a 'daughter' group as it were.
For example there are large segments of the US human population that have historically resisted intermingling due to social circumstances despite the fact that they are inter-fertile, and accordingly I would not expect some alleles not to be evenly distributed through out the population. I suspect we can find examples of animal populations behaving similarly.
Sure but the US is only a couple of centuries old. So any alleles unique to this group are alleles that were not present in the main population of Europe before their ancestors adopted a resistance to intermingling. The alleles might be 'old' relative to say, my shoes. But they are 'new' relative to the human race.
What we can say is that distributing a new allele will take time, and perhaps that means at a minimum that the term "new" is somewhat subjective.
For clarity, 'new' was being used relative to the population.
Also we're talking about perturbations from Hardy-Weinberg which has a large population with non-random mating (ie., anybody in the population has the same probabilities to have sex with any other member of population as any other member of the population). This is to isolate the effects we're talking about.
Faith believes that that a subgroup with different allele frequencies becomes isolated, and because of their allele frequency differences they are morphologically different and that over time these differences exacerbate and by {magic} the two populations are eventually unable to successfully interbreed.
Of course in the complete HW scenario, the original population is infinitely large and so every possible combination of alleles exists in the original population. Thus the original dog population, were it to be infinite, would contain all possible breeds that don't require mutations to express them (according to Faith therefore, all breeds). So the daughter population, if it somehow manage to be selective in which animals went to join it, will only contain a subset of the animals in the original population. So even on the face of it, this cannot lead to speciation in the sense of the two species are genetically incompatible.
The argument seems to be trying to both have its cake and eat it.

This message is a reply to:
 Message 80 by NoNukes, posted 05-28-2016 3:23 PM NoNukes has replied

Replies to this message:
 Message 97 by NoNukes, posted 05-30-2016 1:14 PM Modulous has seen this message but not replied

  
Newer Topic | Older Topic
Jump to:


Copyright 2001-2023 by EvC Forum, All Rights Reserved

™ Version 4.2
Innovative software from Qwixotic © 2024