New blogs

Leherensuge was replaced in October 2010 by two new blogs: For what they were... we are and For what we are... they will be. Check them out.

Tuesday, July 1, 2008

Fundamental conceps of Population Genetics. Abracadabra?


No! Basic understanding. Understanding not even of genetics but of probability.


This is the "official" continuation of a discussion TerryT and I have been involved in, lately totally off-topic-ly in this other thread. Terry said:

Abracadabra, bottleneck, drift, founder effect. With such magic words anything becomes possible.

(...)

... diversification indicates no bottlenecks, founder effect or drift, very unlikely in a population confined to just one region.

What I mean to prove here is that diversification in typical Paleolithic conditions (very low population density, relatively frequent colonization of new lands, specially in the late MP and early UP of Eurasia), necesarily implies drift and founder effect. Bottlenecks instead are a non-normal extreme event that might or might not have happened.

For that I will call as witness the source of all knowldge: God - err... I mean: Wikipedia.

1. Drift: In population genetics, genetic drift (or more precisely allelic drift) is the evolutionary process of change in the allele frequencies (or gene frequencies) of a population from one generation to the next due to the phenomena of probability in which purely chance events determine which alleles (variants of a gene) within a reproductive population will be carried forward while others disappear.

By definition, genetic drift has no preferred direction, but due to the volatility stochastic processes create in small reproducing populations, there is a tendency within small populations towards homozygosity of a particular allele, such that over time the allele will either disappear or become universal throughout the population.

This eventual outcome of drift, when an allele displaces all others within a population, is normally refered to as fixation. Fixation may also happen because of non-random processes such as selection.

2. Founder effect: In population genetics, the founder effect refers to the loss of genetic variation when a new colony is established by a very small number of individuals from a larger population.


Illustration of different possible founder effects from a simplified hypothetical population with two alleles

In addition to founder effects, the new population is often a very small population and so shows increased sensitivity to genetic drift, an increase in inbreeding, and relatively low genetic variation.

I say (and I think most will agree) that these two elements were fundamental in Paleolithic genetics, very specially in the colonization of Eurasia and related lands (Sahul, America). They are no abracadabra: just business as usual.

One could argue that, after Neolithic, the effects of drift have become negligible - and that's certainly the case of modern huge and highly intercommunicated populations. But that was not the case in the Paleolithic, when small bands of hunter-gatherers made up that human reality. Therefore understanding these concepts is essential to understand the formation of Humankind, very specially on light of haploid genetics.

The third concept that Terry likes to mix with the others like if these were the same thing is bottleneck. It is clearly not the same: a bottleneck is a extreme (not normal) phenomenon, where a population is not decimated but almost annihilated.

A population bottleneck (or genetic bottleneck) is an evolutionary event in which a significant percentage of a population or species is killed or otherwise prevented from reproducing, and the population is reduced by 50% or more, often by several orders of magnitude.

This is a very rare happening, though founder effects can appear like population bottlenecks because their effects are similar.

Population bottlenecks have been detected apparently in Eurasians (but not in Africans), slightly sharper among East Asians than among Europeans. This may indicate a real bottleneck (near-extinction) for instance after an event of massive influence like the Toba explosion but it is also possible that the apparent bottlenecks in Eurasians mean just founder effects.

__________

Caveat: I may have misrepresented Terry's views in this post. Please read the comments for his explanations. Most importantly:

As I hope I've made obvious elsewhere I totally accept drift and founder effect (as well as bottlenecks to varying extents) have been important during our evolution. What I am totally uncomfortable with is where they are invoked to explain away inconvenient facts that conflict with particular theories about our evolution.

37 comments:

terryt said...

Maju. As I hope I've made obvious elsewhere I totally accept drift and founder effect (as well as bottlenecks to varying extents) have been important during our evolution. What I am totally uncomfortable with is where they are invoked to explain away inconvenient facts that conflict with particular theories about our evolution. I believe that where the facts don't fit the theory the theory should be altered rather than conjure up unlikely extremes for the operation of these three evolutionary factors. I maintain that they are introduced far more often, and at far more extreme levels, than are justified by the evidence. I'd go so far as to suggest it's possible to prove almost any theory about our origin if you conjure up great enough levels of bottleneck, drift and founder effect. It's almost as magic as abracadabra.

terryt said...

And at the other site you wrote, "I call for founder effects certainly because each time a new land is colonized a founder effect happens". But you call for founder effects from haplogroups that appear not to have even existed in the original region.

If any population is confined to any single region for as long as 50,000 years, as you seem to maintain was the case in the Indian subcontinent, drift alone would narrow the diversity of haplogroups. Yet you maintain that in this region over that time they diversified incredibly. As I've said before, "doesn't make sense".

Maju said...

Maju. As I hope I've made obvious elsewhere I totally accept drift and founder effect (as well as bottlenecks to varying extents) have been important during our evolution. What I am totally uncomfortable with is where they are invoked to explain away inconvenient facts that conflict with particular theories about our evolution... I maintain that they are introduced far more often, and at far more extreme levels, than are justified by the evidence. I'd go so far as to suggest it's possible to prove almost any theory about our origin if you conjure up great enough levels of bottleneck, drift and founder effect. It's almost as magic as abracadabra.

Ok, this is an important caveat.

Nevertheless, it's very different when you have to advocate for extreme succesive bottlenecks, like with the guy that claims a human origin in America (against all evidence and logic but his own kinship systems theory) than when you just have to suggest normal dift and founder effect.

I also understand that, for the model of a West Asian urheimat, one has to call for at least one massive bottleneck too. Not that it could not happen (it's not half of half as incredible as Dziebel's hypothesis) just that there is no other way.

That's the main reason why the rapid coastal migration model is the most popular: because it avoids bottlenecks (other than in South Arabia, if anything) and explains multiple founder effects better.

... you call for founder effects from haplogroups that appear not to have even existed in the original region.

Drift, hehe!

Seriously: you have what you have (the modern haploid reality) and you need to find a most likely explanation (or explanations).

The most clear case for your caveat, I think is Y-DNA D. But Y-DNA is also more susceptible than mtDNA to such strange odds happening and there is some YAP+ (DE undefined, probably D*) in India anyhow.

With mtDNA all can be explained with South Asia as main stop. South Asia has just enough high level diversity to explain all.

If I don't recall badly your main problem here is that N(xR) is not sufficiently diverse in the subcontinent - but there is no alternative unless you want to call for a bottleneck. You certainly present no model that requires no bottlenecks either. There is no such possibility.

If any population is confined to any single region for as long as 50,000 years, as you seem to maintain was the case in the Indian subcontinent, drift alone would narrow the diversity of haplogroups.

Maybe my alternative model doesn't hold. Ok.

But it's not for the reasons you argue. Let's see, hypothetically:

1. Humans arrive to India after a brief and not really sucessful stance in West Asia (either South Arabia or the Fertile Crescent, or both). In my hypothesis this happened before 100,000 BP, in the rapid coastal migration this happened much later: c. 60,000 BP.

2. Eurasians expand in South Asia (not too fast, not too slow). The first effect is diversification of haplos, then localized homogeneization by drift may have happened. The haplogroups lost in this process cannot be documented.

3. Toba event: possible botleneck. If any event can explain a bottleneck, this is Toba. It's possible that this substantially altered the genotype of early Eurasians on random basis. Again, whatever was totally lost cannot be documented. But what survived in small ammounts is surely still around (example: Y-DNA D).

4. Re-expansion within South Asia. Y-DNA K and mtDNA R become important (in separate areas maybe).

5. Expansion out of South Asia (since c. 60-50,000 BP). Clearly the haploid genetics of the groups migrating eastward and westward are somewhat different: the ones migrating eastward are more diverse, the ones migrating westward less so.

This model does not really need a bottleneck but it may include one (Toba). The rapid coastal migration model definitively needs no bottleneck at all other than the apparent one caused by the succesive founder effects.

My alternative model would surely need a "greater South Asia", including parts of SE Asia and West Asia to explain all well. It's by the moment too tentative and immature. I'd need to refine it to be able to provide a detailed explanation, so your criticisms are welcome and are surely of help. If I have to discard it, no problem: the rapid coastal migration is certainly a model I can live with.

terryt said...

"My alternative model would surely need a 'greater South Asia', including parts of SE Asia and West Asia to explain all well". We're certainly moving closer together here. Although I'd expand the region still further.

I actually believe the modern distribution of mt line N and Y- lines C and D is best explained through an early movement east through the gap between the Tien Shan and Altai Mountains. Now, before you say it was too cold, your extended dates for Y-chromosome origin make it much more possible. From 120,000 until 70,000 years ago the climate was easily as warm as, if not warmer than, today. Neanderthal remains dating to more than 100,000 years have been identified from as far north as the Altai Mountains. And interestingly these remains were identified as Neanderthal only by examining their mtDNA. They could have passed for primitive moderns if not for this evidence. Anyway this indicates that before 70,000 years modern humans could easily have moved into the same region.

Climate cooling, perhaps as a result of Toba, later pushed members of the modern human haplogroups south (bottleneck?). This had the effect of basically isolating the above haplogroups to Eastern Eurasia. C5 became the exception, probably developed (by drift) in what is today Afghanistan and Pakistan.

Y-chromosome C4 must have reached Australia during lowered sea level (again a product of later drift on that continent). The most likely period for the arrival is between 70K and 50K. It appears certain that C was in Australia before K arrived anywhere near there. The coastal migration then becomes an East Asian phenomenon, not South Asian, with C* arriving in India from SE Asia. Mt line N tagged along with Y-C although the last stage involved her descendant line R.

Make sense?

Maju said...

From 120,000 until 70,000 years ago the climate was easily as warm as, if not warmer than, today.

It's more like 130-100,000 BP, I understand. IMO it's more like the window to favor the migration from Africa than odd circular routes around the most arid and cold areas these tropical accustomed people could find.

That N* went all the way around northern Asia to migrate to Australia surpasses what I can think as reasonable, really.

Neanderthal remains dating to more than 100,000 years have been identified from as far north as the Altai Mountains. And interestingly these remains were identified as Neanderthal only by examining their mtDNA. They could have passed for primitive moderns if not for this evidence. Anyway this indicates that before 70,000 years modern humans could easily have moved into the same region.

To me it means that it had an owner. Neanderthals were, in principle, much better adapted for subartic climate.

Climate cooling, perhaps as a result of Toba, later pushed members of the modern human haplogroups south (bottleneck?). This had the effect of basically isolating the above haplogroups to Eastern Eurasia. C5 became the exception, probably developed (by drift) in what is today Afghanistan and Pakistan.

Too complicated (my Occam's razor would rather think not). And anyhow C5 is more like Indian than Pakistani, you know.

It appears certain that C was in Australia before K arrived anywhere near there.

Not if K spread since c. 70 KYBP and not if we stay reasonably conservative re. Australian colonization dates (50,000 at most).

The coastal migration then becomes an East Asian phenomenon, not South Asian, with C* arriving in India from SE Asia. Mt line N tagged along with Y-C although the last stage involved her descendant line R.

Again too unnecesarily complicated. And I really do not understand your desire of skipping South Asia in such strange, rather forced, manners when everybody else is suggesting it played a central role (not marginal at all in any case) in the colonization of Eurasia and periphery.

terryt said...

I guess we will just have to await further research. I couple of things I would bet on though.

Firstly, the sort of movement you accept has been happening over the last 50,000 years will prove to be simply a continuation of patterns established long before then, even as long ago as our separation from gorilla and chimp ancestors, or even from orangutan and gibbon. That's how evolution usually works, separation and subsequent recombination of subspecies.

Secondly, humanity did not start suddenly with the appearance of modern Y- and mtDNA haplotypes, even outside Africa. In other words there was no moment when we "suddenly became human". The possession of modern haplogroups does not automatically mean modern-looking phenotype and primitive phenotype does not necessarily eliminate the possibilty of modern haplotypes.

Our evolution was nowhere near as simple as is being promoted by most scientists.

terryt said...

And I've just found this:

http://scienceblogs.com/afarensis/2008/07/03/the_more_things_change_the_mor/#more

Maju said...

I guess we will just have to await further research. I couple of things I would bet on though.

Sure: we are always waiting for future research. To improve our knowledge, to challenge it... Nothing is definitive or absolute.

That's how evolution usually works, separation and subsequent recombination of subspecies.

You mean speciation. Isolation, either physical or bio-cultural (or both) is the main mechanism. Sure that introgression of adaptataive genes happens but, while interesting, it doesn't change that much about the overall process. It's just the cherry, not the cake.

Secondly, humanity did not start suddenly with the appearance of modern Y- and mtDNA haplotypes, even outside Africa. In other words there was no moment when we "suddenly became human".

Sure. It's just a convenient reference from the viewpont of genetics. But the case is that we did become human (in the sense of Homo sp. first, and in the sense of Homo sapiens later). Arguing about the exact moment(s) is interesting but it certainly happened in Africa some time after the branch leading to Neanderthals broke apart. While the genetic age estimates for Neander-Sapiens diveregence are of "only" 500,000 years, the fossil record, if I'm not wrong, is almost double that time. Lions and tigers have been separated for the same period roughly and they are just barely able to produce hybrid offspring (always in captivity, never in the wild).

The possession of modern haplogroups does not automatically mean modern-looking phenotype and primitive phenotype does not necessarily eliminate the possibilty of modern haplotypes.

Not sure what you mean with this. Nor how to answer. Modern means H. sapiens or refined 20th-century like humans? There were certainly archaic-looking humans in our genealogical tree: many of them. And one could well argue that some groups and individuals look rather archaic even today. But are you talking of human phenotypical variation or of different species?

And I've just found this:

http://scienceblogs.com/afarensis/2008/07/03/the_more_things_change_the_mor/#more


Hmmm... He's talking about a 1985 (!!!) paper that claims that human mtDNA is actually Chimpanzee mtDNA (mtDNA does not recombine!). There's no way to see hybridation in that because it's a yes/no switch: there are no greys, no rainbow, just black and white, this or that. There is no possible talk of introgression or admixture in haploid genetics: it's mere maternal or paternal ancestry. If mtDNA Eve had a Chimp haplogroup (what is obviously not the case, don't believe anything published before 2000) it would not matter at all: she would still be our common ancestral great-great-...-grandmother (by purely female lineage).

In brief: there is a very good chance (at least as far as the state of the art of genetics can confirm) that H. sapiens never mixed significatively with other species since mtDNA Eve. Possible introgression excluded, of course, but that is not significative admixture, just anecdote.

I know this is challenging for some people's beliefs but it seems what all evidence supports so far.

Treating H. sapiens as a totally separate population for nearly all purposes makes a lot of sense in any case.

terryt said...

"You mean speciation". No, I don't. Evolution is entirely possible without speciiation.

"Isolation, either physical or bio-cultural (or both) is the main mechanism". Inbreeding is the main mechanism. It may in turn be a product of either form of isolation of course.

"Sure that introgression of adaptataive genes happens but, while interesting, it doesn't change that much about the overall process". Introgression has been extremely common during the evolution of many species, and is possibly one of the main methods of genetic change in populations.

I understand lions separated from tigers more like two million years ago. Horses, donkeys and zebras about three. Elephants a little more than that again. I realise many people like to see humans as being a special case, and perhaps they are.

Maju said...

I think I must agree with you re. your last post.

Isolation and inbreeding: ok, they are about the same.

Introgression may be important. Sure. It just doesn't account for the bulk of the heritage but for single very adaptative genes. The overall genome doesn't really change and the same result could have been achieved via a mutation.

If red squirrels send specific super-adaptative genes to grey ones via introgression, that would probably just speed up the extinction of the first ones. Grey squirrels would still be that, not really hybrids, but also would have gained in fitness re. to local enviroment, so their success would be even greater.

It's more like sort of natural "genetic engineering" than admixture in the usual sense.

I understand lions separated from tigers more like two million years ago.

I was talking from memory, so maybe you are right about that.

I am not sure there was any major biological barrier to Neander-Sapiens hybridation, except maybe birthgiving or complications. But the case is that it doesn't seem to have been widespread, so surely cultural and identitary elements acted there at least as much as the biological ones, probably even more.

terryt said...

"Introgression may be important. Sure. It just doesn't account for the bulk of the heritage but for single very adaptative genes".

I was actually thinking of how single genes, especially recessive genes, move through a population. They must first become fixed in some subset of the population through inbreeding. After all selection can operate only on expressed genes. Advantageous recessives would then spread through the wider population in a manner reminiscent of introgression, although technically the term is reserved for iter-species gene transfer. Perhaps in such a case we could regard the gene movement between such populations as being that between subspecies.

terryt said...

I just thought I'd explain something you perhaps misunderstood in a link I provided a few days ago. You wrote,"There is no possible talk of introgression or admixture in haploid genetics". But that's not what the author was saying. The idea was that although Australopithecus is a human ancestor the evidence of a more recent chimp/human split could be explained if mtDNA had introgressed into Australopithecus more recently than the original chimp/human split.

"this dating may pose a problem for the widely believed hypothesis that the bipedal creature Australopithecus afarensis, which lived some 3.7 million years ago at Laetoli in Tanzania and at Hadar in Ethiopia, was ancestral to man and evolved after the human-ape splitting. Another likelier possibility is that mtDNA was transferred through hybridization between a proto-human and a protochimpanzee after the former had developed bipedalism".

There is no hint in the paper that the author believes mtDNA recombines (even though "He's talking about a 1985 (!!!) paper").

Maju said...

I was actually thinking of how single genes, especially recessive genes, move through a population. They must first become fixed in some subset of the population through inbreeding. After all selection can operate only on expressed genes. Advantageous recessives would then spread through the wider population in a manner reminiscent of introgression, although technically the term is reserved for iter-species gene transfer. Perhaps in such a case we could regard the gene movement between such populations as being that between subspecies.

Not sure what to make with this. normally recessive genes are disadvantageous or neutral.

Guess we can use the term introgression for some particular type of gene-flow between same species' populations that are basically isolated from each other but I suspect you are abusing it.

Introgression basically means that population X transfers gradually, through many generations, an adaptative gene A to population Y. There is a diffuse population Z that acts as intermediary. This Z hybrid population is generally very small and marginal (otherwise it would not be introgression but hybridation). Non-adaptative genes have it very difficult to flow through introgression, much more deleterious ones. But those that are clearly adaptative have Darwinian selection on their side and that often overcomes the strong odds against neutral or maladaptative introgression. It needs time, much time, in any case.

You wrote,"There is no possible talk of introgression or admixture in haploid genetics". But that's not what the author was saying. The idea was that although Australopithecus is a human ancestor the evidence of a more recent chimp/human split could be explained if mtDNA had introgressed into Australopithecus more recently than the original chimp/human split.

That is what the author is saying: he/she is using the term "introgression". Furthermore he's using it in the haplopid realiy. That is not really acceptable: all odds are against.

In brief: mtDNA does not introgress. At least it's not supposed to.

To get a haplogroup fixated (specially mtDNA, that suffers much less sexual bias than Y-DNA) it must be a very common one in the original (pre-fixation) population. That is precisely what you should not find in the case of a very minor hybrid group, maybe a single individual.

Introgression happens against normal odds because the introgressed gene is clearly adaptative. Otherwise it just doesn't happen. And certainly a single adaptative gene (allele) can only do that undavertedly inside recombined chromosomes. Slowly, gradually, it becomes the only imported piece of DNA in the recieveing population and, after that, it becomes dominant and even fixated through selection.

This process can't happen with haploid DNA. If australopitecne mtDNA is Chimp one, then Austrlopitecines would be largely Chimp: a hybrid.

This is obviously not the case.

There is a much more simple explanation: TRMCA estimates are just mere educated guesses and must be corrected in agreement with the fossil record, not the other way around. You can't found a theory only on genetic age estimates because it's like founding astronomy only on stellar magnitudes or the apparent movement of the Sun around Earth. If there are other facts that contradict these guesses then they must be reviewed.

Genetics can show quite clear philogenies but it's much worse at estimating timelines.

terryt said...

"normally recessive genes are disadvantageous or neutral". But obviously not always so or evolution would consist entirely of eliminating disadvantageous genes. Most genes, including adaptive genes, usually first arise as recessives.

Advantageous recessive genes flow between populations of the same species but the difference between a population, a subspecies or a species is a matter of degree. Gene flow between populations is usually easier than that between species but all three represent stages on a continuum. Although the term introgression is technically confined to just the last of the above three stages of difference, hybridism and introgression are part of the same continuum.

We can see that gene flow even between populations is far from unimpeded. That's why most species, including humans, demonstrate regional variation.

Maju said...

Although the term introgression is technically confined to just the last of the above three stages of difference, hybridism and introgression are part of the same continuum.

Well, I'd say introgression is not exactly but almost the opposite of hybridation. If 50-50 hybridation is white, introgression has so little white that is black for all practical purposes.

It's a drop versus the sea, a grain of sand versus the vastness of the desert, the density of matter in outer space versus the density in a neutron star. Sure: pure void is nowhere but introgression is almost that.

terryt said...

Maju. That is totally stupid.

Maju said...

Thanks for your appreciation. May I know why you think it is stupid?

How many multiallelic genes does the human genome have? How many genes are involved in a typical introgression? I'm not sure right now about the first answer, just that it's a very high number, but I'm pretty sure about the second: one.

Introgression is one drop in the ocean... or at least in a very large pool.

Maybe for you it's "stupid" but for me it's just the plain truth.

terryt said...

"I'd say introgression is not exactly but almost the opposite of hybridation". Opposite ends of a spectrum more likely. There is no reason at all why hydridation events should necessarily be 50:50 nor why introgression should involve just one gene from a single hybrid individual.

You seem to believe that species, subspecies and populations are fixed entities, perhaps formed during a creation event?

You're no doubt familiar with the statement Miguel de Unamuno is said to have made, "In Spain even the atheists are Catholic". The belief you seem to hold is an example of this survival. Species, subspecies and populations are continually evolving entities, continually exchanging genes.

To understand how it works you might like to check out my essay "Chromosomes and DNA" and scroll down to 'Dominant and Recessive Genes' and then look at "Hybrid Vigour and Inbreeding" and scroll down to 'Wave Theory of Evolution'.

Maju said...

You're no doubt familiar with the statement Miguel de Unamuno is said to have made, "In Spain even the atheists are Catholic".

No, I'm not. I never liked that guy. He was fascist and pro-Spanish and his literature is kinda boring for my taste (as most Spanish literature, btw).

You seem to believe that species, subspecies and populations are fixed entities, perhaps formed during a creation event?

You seem to read my mind... in a totally wrong way.

No. I don't believe they are "fixed". I think species evolve continuously, mostly through random mutations and selection (and merely random drift) within their internal diversity.

Subspecies and populations are a different category of things. They are obviously much more open to outbreeding within their own species.

"I'd say introgression is not exactly but almost the opposite of hybridation". Opposite ends of a spectrum more likely. There is no reason at all why hydridation events should necessarily be 50:50 nor why introgression should involve just one gene from a single hybrid individual.

Opposite sides of a spectrum certainly.

Of course, I used that example of 50:50 admixture as one extreme but even 10:90 admixture is still quite "white" in comparison with a single introgressed gene. 10% of the whole multiallelic part of the genome is still many thousands genes. It's a totally different category than introgresion, where a single gene has been selected after extensive filtering via real hybrids (that may even have gone extinct after they fulfilled their role).

Anyhow, just for the record, even if Neanderthal introgression is possible it's not yet demonstrated as real.

terryt said...

"I think species evolve continuously, mostly through random mutations and selection". So how exactly do these 'random mutations' actually move through the species? Surely by forming hybrids with those individuals in the population that don't have the mutation.

In effect each "single gene has been selected after extensive filtering via real hybrids", as opposed to unreal ones?

You seem to greatly exaggerate the distinction between species and subspecies. If the distinction is as great as you believe how come there is so much disagreement within biologists as to whether certain populations represent a separate species or just subspecies?

Maju said...

So how exactly do these 'random mutations' actually move through the species?

Do they? Certainly they may move (via introgresion) but it's not a process "by default".

The "by default" process is internal evolutions of species (and sometimes of subspecies: speciation) separately.

In effect each "single gene has been selected after extensive filtering via real hybrids", as opposed to unreal ones?

Yes. As opposed to non-hybrids with one or two possibly introgressed genes.

I would never call those inidividuals "hybrids" but you seem to want very much to do it, so the difference between a 50:50 or 30:70 (real) hybrid and a virtually "purebreed" individual that has one introgressed gene gets blurred, so you can keep mixing apples and oranges in your stubborn attempt to persuade yourself and the rest that H. sapiens (or some populations of this species, I assume you exclude Africans from that) is "mixed" with Neanderthals in a significative ammount.

No offense meant but I think that's what you are trying to do, consciously or unconsciously: to supress the difference between wet and dry so you can claim that the platter is wet even if it just has a single molecule, not even a drop, of H20.

I cannot accept it, sorry.

You seem to greatly exaggerate the distinction between species and subspecies. If the distinction is as great as you believe how come there is so much disagreement within biologists as to whether certain populations represent a separate species or just subspecies?

I don't think I want to exaggerate the differences between species. I even do admit that the two human species might hypotetically have mixed and even that such thing probably happened, in spite of the feeble evidence in support of it.

But I also see the evidence, both forsenic (archaeological) and genetic and I realize that it must have been something mostly inconsequential in the big picture. That overall, most Sapiens did not actually mix with Neanders, at least in a way that left viable lineages.

Why? I can't say for sure: it may have been biological reasons or cultural ones, of the same type that got the Neanders wiped out, while our ancestors survived. The case is that we don't see any conclusive evidence (even at the minimalistic level of introgression) so far that shows that succesful admixture actually happened.

Lack of evidence doesn't mean that it did not happen for sure but it also doesn't mean the opposite. And as the negative evidence mounts up, the proof of admixture remains as elusive as ever.

Occam's razor.

terryt said...

Maju. How would it be possible for any gene to introgress without the formation of at least one hybrid individual?

And how does "internal evolutions of species" occur without the expansion of particular genes through the geographic range of that species?

Lastly, what is your definition of a hybrid?

Maju said...

Maju. How would it be possible for any gene to introgress without the formation of at least one hybrid individual?

It can't (or maybe it can thanks to viruses and genetic engineering?). Typically there's a small hybrid population at some time and place that acts as mediator between the two affected species.

But the important thing is that once that barrier is overcome the intermdiate population does not need to exist anymore and that, soon, only the adaptative introgressed genes are really surviving (via selection, that counters drift for that particular overly adaptative gene).

And how does "internal evolutions of species" occur without the expansion of particular genes through the geographic range of that species?

The expansion of internal mutations is totally independent of any possible admixture with other sepecies and introgression.

Mutations happen, as you know well: the species does not need to hybridate with other groups to evolve. Its own internal genetic diversity (almost always growing via mutations) makes that possible.

It's so self-evident, so basically 19th century science... that I really can't believe you are asking that.

Lastly, what is your definition of a hybrid?

I don't have any particular definition. Just the standard: an individual that is genetically intermediate between two or more species.

So Cupressocyparis x leylandii is an (artificial) hybrid between a species of Chamaecyparis and another one of Cupressus. It's an interesting hybrid because it blurs not just the barriers between species but also between genus. But it's a plant and plants are more flexible genetically than most animals.

A more standard example is the once common mule: a hybrid of horse and donkey. A horse that has a single donkey gene is not a hybrid, though considering mules are steryle that can only happen via genetic engineering or viral transmission.

terryt said...

"I really can't believe you are asking that".

The reason I asked was because you don't seem to understand how evolution actually works. On the other hand you have finally admitted that "Typically there's a small hybrid population at some time and place" involved in introgression.

"Mutations happen, as you know well: the species does not need to hybridate with other groups to evolve. Its own internal genetic diversity (almost always growing via mutations) makes that possible". And here I was, hoping you'd understand that the mutations do not instantaneously spread through the whole population.

Your definition of 'hybrid' as "an individual that is genetically intermediate between two or more species" is clearly inadequate. In other words you're saying that when farmers use the term 'hybrids' to refer to crossings between different breeds of the same species they are incorrect. Surely a hybrid is simply a product of parents who are genetically different. Of course once a gene mutates in an individual he or she is then genetically different from the remainder of the population, perhaps only in a single recessive gene. So the new gene, by definition, spreads through the population through the formation of hybrids within that population. The statement "The expansion of internal mutations is totally independent of any possible admixture" is therefore totally incorrect. The new gene has to spread by admixture.

By the way hybrids between horse and donkey are not always sterile. In these cases are the particular horse and donkey involved suddenly members of the same species?

Maju said...

Terry: this is beginning to be true nitty-pickiness and I would like to know where you are trying to head the discussion, if to anywhere at all.

But anyhow:

...you have finally admitted that "Typically there's a small hybrid population at some time and place" involved in introgression.

Your tone is accusatory: "you have finally admitted that...". Did you even ask?

Anyhow, what does this imply to you? For me it's almost trivial, meaningless, accidental. I find much more weirdo the insertion in laboratory of tomato mtDNA in maize or whatever.

And here I was, hoping you'd understand that the mutations do not instantaneously spread through the whole population.

And now you are accusing me of saying, or maybe suggesting that? I can't believe it.

When did I say that mutations spread around like that?

Your definition of 'hybrid' as "an individual that is genetically intermediate between two or more species" is clearly inadequate. In other words you're saying that when farmers use the term 'hybrids' to refer to crossings between different breeds of the same species they are incorrect.

In my understanding yes. But guess it can be used loosely.

Can you tell me anyhow how does this matter at all?

Surely a hybrid is simply a product of parents who are genetically different.

Every two non-clonic individuals are genertically different. C'mon! Are you telling me that every single individual incepted through sexual reproduction is a hybrid?

But again, how can this matter? If you want to use the term hybrid to mean every single person, animal or plant, do it please: just make sure that I know what you are talking about.

Of course once a gene mutates in an individual he or she is then genetically different from the remainder of the population, perhaps only in a single recessive gene. So the new gene, by definition, spreads through the population through the formation of hybrids within that population.

I think there are much better terms in the field of biology to refer to that. Anyhow, certainly a single gene does not define a hybrid, it defines a carrier. Even with the farmers' loose definiton of a hybrid, a burmese cat with a mutation that makes it slightly larger or smaller, or more agile maybe, or with a sharper nocturnal vision or whatever... is still as much a burmese cat as the one that doesn't carry it. A hybrid would be the offspring of a burmese and a persian cat, notwithstancing that mutation.

I don't accept you can use the term "hybrid" based on a single gene, specally as every two people, except identical twins, are different in many many genes. It's not just mutations, but also (and specially) recombination.

My brother and I are statistically only 50% the same. Of course many genes are the same between my father and my mother (they are both humans and, just to refine it a little more, Europeans) but many are very different anyhow. These are recombined randomly and so I have black hair while my brother has blond hair, I am some 10 cm taller than him, he is quite stockier than me, etc. We have other things in common though.

That doesn't make any of us "hybrids". I'm just carring some genes he is not and vice versa. That's all.

The statement "The expansion of internal mutations is totally independent of any possible admixture" is therefore totally incorrect. The new gene has to spread by admixture.

If by admixture you mean sex... maybe. But it's not what most people has in mind, specially when discussing the admixture of neanderthals and sapiens. People doesn't think in sex and reproduction between to H. sapiens that are at least 99.9999999999% that beyond doubt. A hybrid in this context is clearly someone who is like 50-50, like a mule or a liger, or maybe 20-80 but not someone who carries a single hypothetical introgressed gene.

Arguing about that will only get us in a circular discussion. Please don't insist. Use the term carrier or whatever you can find that describes that biological situation properly.

Otherwise we are all hybrid of human and viruses, who have contributed to our (junk) genetic pool in great ammounts. And hybrid of our mom and dad. It's crazy terminology.

By the way hybrids between horse and donkey are not always sterile.

Didn't know that.

In these cases are the particular horse and donkey involved suddenly members of the same species?

Obviously not. And?

Maybe you should ask yourself what happens after that fertile mule and its descendants have been living for some 10, 100, 1000, 10,000 generations between horses only. Are their descedants still mules, hybrids?

I say horses. And no one would dare say otherwise, specially because they can't even know. But even if knowing, they would still use the term horse, not hybrid.

terryt said...

"I would like to know where you are trying to head the discussion". This: "Every two non-clonic individuals are genetically different". In other words genes move relatively independently through a population. The particular population may be anywhere between a small inbred group to a group including members classified as being different species, and all points in between. Selection acts on these moving genes and produces evolution. It is a dynamic process. Groups move apart and come together.

"what happens after that fertile mule and its descendants have been living for some 10, 100, 1000, 10,000 generations between horses only". That was my point. Hybrids are absorbed by the majority population but their genes rapidly become widely distributed through that population. Neanderthals were probably outnumbered by incoming modern humans but their genes could well have reached all other parts of the world by now. So how do we identify these genes?

Maju said...

In other words genes move relatively independently through a population.

Relatively indeed: you forget selection and drift.

But the latter strongly tends to favor the most common genes (just probability laws) and to displace the less common ones. Unless selection is at work in the opposite sense.

That's how introgression may happen: the alien genes are basically supressed by mere drift, except for the few that are really adaptative. There is some uncertainty, randomness, in the whole process but essentially it strongly tends to be that way.

The particular population may be anywhere between a small inbred group to a group including members classified as being different species, and all points in between.

Generally they tend to be more like localized inbreeding populations. That's normally how speciation (and differentiation within one species) happen.

Interspecies admixture can happen but it's very rare anyhow. At least in nature. Much more common is admixture within more or less distinct populations of the same species.

The borders of species may have been blurred somewhat by recent discoveries but it's not like they do not exist at all: lions and tigers do not mix in nature, only in zoos. There are ecological, biological and behavioural/cultural barriers between species. Now we know that occasionally these barriers can be overcome but it's not like it's happening every day everywhere. It is very rare, specially in comparison with intras-species "admixture" (i.e. normal reproduction).

Hybrids are absorbed by the majority population but their genes rapidly become widely distributed through that population.

Mostly they vanish. The odds are strongly against them. It's not just drift but also, for most genes, intra-species selection (like sexual selection) will also be acting against. Only those genes that really add a clear fitness value are likely to survive in the long run.

It's not like everything in a inerspecific hybrid is likely to be adaptative within a abtural single-species context. Maybe the mule is stronger than normal horses but also slower, and therefore easier prey for wolves and the like. You assume that just for being a hybrid it's best fit, that the corresponding genes will multiply like rabbits. But that's not the case normally: most probably most genes will just not survive selection trials or the mere randomized drift, as it's one among many.

Neanderthals were probably outnumbered by incoming modern humans but their genes could well have reached all other parts of the world by now. So how do we identify these genes?

You tell me. It's on the proponent that falls the burden of the evidence.

I can just suggest: fully genotype several Neanderthal aDNA (as time passes this possibility is becoming easier) and compare that with modern human genome. Or do it in a fraction of the genome (say chromosome 21 or whatever) to begin with.

It's the only way to prove it. Otherwise it will remain a mere hypothesis forever.

Of course, you may find nothing. So far this has been the only result - but the scope of the tests done so far would not be able to detect minimal introgression. Only expanding the genotyping, fully mapping the Neanderthal genome, or at least a significative fraction of it, can we advance in this question.

But for me it's a minor issue. I don't have strong passions on it and I understand that the evidence so far basically disproves large scale succesful admixture in any case.

Wether Neanders gave us some specific genes it's possible, it would be interesting to know, but it's not really of outmost importance to define the bulk of H. sapiens ancestry, that is intra-specific for some 99% or probably more.

It's not that I don't have any curiosity about possible Neanderthal introgressions. But I really don't care that much either. It's not something that will redefine the species, not at all.

terryt said...

"fully genotype several Neanderthal aDNA ... and compare that with modern human genome". That won't tell us which direction any genes in common came from. Unless we sequence 200,000 year old African aDNA.

"It's not something that will redefine the species". It sort of would, really. Accepting it would let us more simply explain our evolution to evolution-deniers. We would then have evolved in exactly the same manner as every other species on the planet, not somehow different.

Maju said...

That won't tell us which direction any genes in common came from. Unless we sequence 200,000 year old African aDNA.

Maybe. It depends of the specimen. But anyhow, please, at least some indication that there could be anything before we make a huge issue out of it, ok?

Accepting it would let us more simply explain our evolution to evolution-deniers.

I'm absolutely uniterested in what evolution-deniers think: it's unscientifical and therefore only political. Don't force science just to make a political point because then you'll become like them.

We would then have evolved in exactly the same manner as every other species on the planet, not somehow different.

I'm not so sure introgression is sooo common. It's just a "new" idea that has yet to be tested in the majority of living species.

And certainly the vast majority of species if not all, do not evolve via admixture. Maybe introgression plays a role now and then but true "hybrid species", if they exist at all, must be the bearded woman of biology, not the norm.

I only remember one case some hoped a hybrid iguana might be the origin of a new species and... it was sterile. It is much more likely (and "classical") that a new species of iguana evolves from an isolated group in a remote island than via hybridation. That's how species normally evolve: speciation by isolation and (mostly) inbreeding.

If there is no inbreeding there can hardly be any speciation.

terryt said...

"And certainly the vast majority of species if not all, do not evolve via admixture'. Wrong.

Two populations do not wake up one morning and suddenly decide to be separate species. There is ample evidence of mixing in the early stages of speciation. Cattle (and most interestingly: Bison), mallard ducks, the chimp/human split. If you're still not convinced I'll find the articles and other examples but you seem totally entrenched in your single origin of each species theory.

I agree that "If there is no inbreeding there can hardly be any speciation'. But if there is ONLY inbreeding extinction, rather than evolution, is a much more likely outcome.

Maju said...

Two populations do not wake up one morning and suddenly decide to be separate species.

Obviously not. It's not a conscious process.

There is ample evidence of mixing in the early stages of speciation.

Maybe. It will depend much of the specific circumstances of the process. It doesn't matter much anyhow: if Pan and Homo would have remained mixing their genetic pools at regular rates since the moment of their historical split there would have been no split at all.

At some time the two species (now genus, somewhat arbitrarily) were occupying different spaces (jungle and savannah probably) and doing different everyday things... and reproducing mostly only with their own group.

I agree that "If there is no inbreeding there can hardly be any speciation'. But if there is ONLY inbreeding extinction, rather than evolution, is a much more likely outcome.

It depends much of the variety of the genetic pool of the inbreeding group, what basically means the size of the inbreeding population. Inbreeding is very common though usually kept (by nature or custom) at some tolerable levels (that's why we forbid incest but do not care so much if cousins marry occasionally, much less if they are distant cousins). If two population of, say, several thousands individuals are inbreeding for many generations, they will tend to diverge more and more and will eventually become separate species.

If in the middle of this process some occasional admixture happens, it will not substantially alterate the process, it's almost trivial. If a lot of admixture happen instead, the speciation process may be stopped (both groups become one). Evolution instead continues all the time through mutations, recombination, selection and even neutral drift.

terryt said...

"Evolution instead continues all the time through mutations, recombination, selection and even neutral drift".

Maju we're getting somewhere. you've actually included 'recombinations'. But you have left out 'inbreeding'.

And I notice elsewhere you are now conceding that haplogroups can be replaced.

terryt said...

Just to take you up on another aspect. You wrote earlier, "I'm absolutely uniterested in what evolution-deniers think".

Unfortunately these evolution-deniers are still extremely powerful and influential. Mythical views of history are continually being used to justify exploitation and oppression. I believe it's important we expose these myths.

Maju said...

Maju we're getting somewhere. you've actually included 'recombinations'. But you have left out 'inbreeding'.

I did not. That's only if you take that sentence out of context. If you read the whole section of the reply, the concept of inbreeding is already mentioned and insisted upon. What that means is that with inbreeding and/or outbreeding, with speciation or without it, evolution never stops because of those reasons anyhow.

And I notice elsewhere you are now conceding that haplogroups can be replaced.

They can, though for a total replacement it's needed a lot of drift, and therefore very small populations. It's much more likely to happen in the Y-DNA than in the mtDNA side anyhow, if at all, for obvious reasons.

Just to take you up on another aspect. You wrote earlier, "I'm absolutely uniterested in what evolution-deniers think".

Unfortunately these evolution-deniers are still extremely powerful and influential. Mythical views of history are continually being used to justify exploitation and oppression. I believe it's important we expose these myths.


Well, we don't have such stuff over here. It's a Gringo or Anglo phenomenon mostly, like crop circles. The only people I know in RL who dare to reject evolution are ignorant and uninfluential hotheads - and they are extremely rare anyhow. There's some people who want to believe some sort of teleology (subtle divine direction of the natural process) but they don't really challenge science, just they seem to need to integrate the divine into it somehow according to their pesonal beliefs.

In any case I think it's a political matter: their superstitons and pseudo-science need to be denounced, of course, but for that you do not need to deform real science. If you do you'd be falling as low as them and may even fire back.

Science is indepedent of what some nutty hotheads believe, or at least should be.

I think that falling into their game is giving them too much credit. Their maneouvres and falsehood should be rejected politely and firmly but one should not let the rivals to set up the agenda. Never.

Personally I think that when dealing with fanatics, it's best to question the very basis of their beliefs. For example, as for scientific evidence of the existence of "God" or suggest that they are book idolaters and ignoring the real message of "God" in reality. It's best to attack them rather than just stay on the defensive. If they are busy trying tod demonstrate that "God" exists, or that their holy book is the true word of "God" in opposition to say the Satanic Verses or whatever, or they have to fence off a campaign to deprive them from their tax privileges... then they won't have time to bother you with their utter nonsense.

Thanks God, it's good to be anticlerical and somewhat mischievous. Jehovah witnesses and Mormons flee from your home, you can even enjoy trying to convert them to atheism, paganism or even satanism, if you are in the mood. Never ever let those jerks take the initiative. 'Hasta la victoria siempre!'

Enjoy.

Maju said...

PS - I learned all that in a Jesuitic school. No priest could debate with me: they just don't have arguments, as all their belief is just based on mere faith (illusions). I made of extremely boring religion classes a rather enjoyable battlefield on wether miracles were real or mythical, or if Jesus was an ET... whatever I felt like.

Finally, after some struggle, I managed to get out of that damn prision.

Always take the intitiative, always get them on the defensive. It's easy: they cannot win because they have nothing to support them. It's not a battle between science and creationism, it's a battle between common sense and illusory faith. Abnd it's soooo easy to offend them and get them more or less violent while you just enjoy a good laugh...

:-D

terryt said...

We obviously do agree on much more than we disagree on. I too "enjoy trying to convert them to atheism, paganism or even satanism, if you are in the mood".

I was actually thinking of the situation in Israel/Palestine with my original comment regarding evoluion-deniers. The idea that Jews are somehow different from their neighbours is based totally on myth. The support for George W. Bush's 'war on terror' comes mainly from those who believe in the literal truth of certain myths.

Anyway, I think we can leave the subject now. As I've said, I totally bow to your knowledge of more recent (more recent than 40,000 years that is) movement of Y and mtDNA haplogroups. Thanks for your contribution to my question at Dienekes.

Maju said...

Yes, we do agree in a lot of things, most of the discusion is on emphasis and some matters of perspective.

And I also agree that the Palestinian coflict is largely based upon myths. For example many pro-Zionists seem to believe now that Jews were expelled by the Arabs in the 7th century, what is totally false. They were genocided by the Romans in the 1st century instead and what the Arabs found was an Eastern Roman mostly Christian province instead. Diaspora Jews were already in the diaspora before that epysode anyhow. Also they conveniently forget about the very conquest of Canaan by Jews only some 1000 years before that Roman genocide.

And yes, we have to close this discussion at some point. I am sure we will get to greter extents eventually anyhow.

Enjoy.