Posts
The Genomic Timeline of Becoming Homo sapiens
12 hours ago
Friday, April 10, 2009
Note: preliminary mtDNA timeline
After squeezing my neurones for many days I have found a reasonable (albeit still hypothetical) timeline for Eurasian mtDNA based on SNP count. I have used a very simplified tree/sequence, that focuses in West Eurasian lineages, as these are the easiest ones to match with the known archaeo-cultural record, at least for me.
I used the mtDNA tree at PhyloTree.org as reference because it seems best kept and updated.
L3>>>>M>N>>R>>R0>HV/U>>H/U6/U8>>>U5>>>>U5a>>K>T/W>K1/K2>>>>>>>>"present"
Obviously this sequence is not a genealogical tree but a mere lineal count of some representative nodes (epysodes) in the Eurasian genetic tree. The different lineages have been blended in a single "temporal" line for working purposes only. Each ">" symbol represents 1SNP in relation to the previous node and to the general L3 node (that includes all Eurasian plus some African mtDNA).
Of course, my assumption is that each SNP should represent an approximately similar period of time in the evolution of the Eurasian mtDNA.
"Present" is considered here by taking some well studied K subclade. I decided to use it after preliminary attempts made modern K sublineages to be well "in the future". At this stage it has not be contrasted with other lineages, what may produce some relevant fine tuning later on.
I tried hard (I swear) to make the H "explosion" fit with Gravettian but that only pushed R0 and U too late into the UP, and made the M,N,R "supernode" to be also incredibly too recent. Also contemporary U6 and U8 are way too likely to be of Aurignacian (Dabban in the case of U6) times. So the H/U6/U8 is here fixed to 41,000 BP, being the other reference point.
The results are (rounded to the nearest milennium):
·L3 67 kya
·M 59 kya
·N 56 kya [9 M sublineages]
·R 52 kya [D, M14, M30, M36]
·R0 48 kya [G, M8, M43, M45]
·U/HV 45 kya [CZ, M31, M34, M39, M48, N2, B, R9]
·H/U6/U8 41 kya (assumed) [Q, N9a, N9b, U2, HV0, R5, R8]
·U5 34 kya [M1, M28, U6b'd, U8b'K]
·U5a 26 kya [M32, R0a,U6a, U7]
·K 21 kya [M21, M22, U4]
·T/W 19 kya [N2a, R11, U6c]
·K1/K2 17 kya
Likely contemporary nodes not directly considered are listed in [brackets] for reference. Not all other major lineages/nodes are coincident with the ones considered here; the ones listed are just for a quick reference of the wider Eurasian picture.
Notice that D and G are listed among the oldest East Asian nodes to branch and they are also the ones considered oldest in Central Asia (derived subclades, not considered), what suggests that the oldest colonization of this region had an Oriental origin.
Overall the impression is that:
1. The "fast coastal migration" scenario, via Bab el-Mandeb, seems ratified, as the dates could not fit with an older Eurasian coalescence. Would I try to force that, the H "explosion", for instance, would have to be much older than any European AMH presence, what makes absolutely no sense. This seems to contradict my previous estimates for Y-DNA, so I will have to review them.
2. When the R "explosion" happened, there were already N-lineage people in East Asia (N9) and Australia (S), and most M sublineages (16/26) had already began branching out in their respective destinations (exceptions include the only West Eurasian M sublineages: M1 and M48). This probably means that the expansion of R implies a second distinct pulse of the Eurasian expansion of AMH, I guess that associated with that of Y-DNA K.
3. AMHs arrived to West Asia and expanded there c. 48-45 kya. The later date may imply also the colonization of parts of the Balcans (Bakho-Kiro).
4. The main expansion of West Asians into Europe and North Africa happened c. 41 kya (this is presumed but should pose no problem as we know that well enough on archaeological grounds). H, U6 and U8 formed and began branching out ("exploding" in the case of H) at that time.
5. The timeline of U5 approaches that of Gravettian (original at c. 28 kya) but not well enough. It must be noted anyhow that the western expansion of Gravettian is much more recent, closer to the Solutrean origins, what gives some margin for speculation. Eastern Europe was basically colonized in this period, what explains well why U5 is more common towards the East, while H is towards the West of Europe.
6. K appears to have a Solutreo-Magdalenian time frame, what would mean that U8 as a whole, also U8b'K, probably coalesced in Europe and not in West Asia. At first I suspected that K would end up being Neolithic but, after some reality check (mutations to present), it seems not. West Asian U8b (also found in Italy) and K is therefore product of a back-migration therefore. I am at the moment uncertain but suspecting that T2 might have migrated to Europe also in this prehistorical context.
And for those who worship the TRMCA (i.e. Tandem Repeat Molecular Clock [Hypothesis] Age [Estimate]) model, I suggest them to re-read my favorite Grimm brothers' tale: The Emperor's New Clothes, which is very illustrative of what happens when people trust what they can't see, and when blind acceptance of authority on the fear of being called a fool goes mainstream, even if it's an obvious nonsense. In any case, as this estimate of mine clearly conradicts many of these elaborate guesses (even if it fits surprisingly well with known Prehistory), I'll be glad to discuss their concerns in the comments section, like always.
.
Subscribe to:
Post Comments (Atom)
73 comments:
"When the R 'explosion' happened, there were already N-lineage people in East Asia (N9) and Australia (S)".
I've been telling you that for ages, but you've only just now come to agree with me. Congratulations. Of course we'd still argue over the route the N-lineage took to reach East Asia and Australia but we'll leave that for another day.
"This probably means that the expansion of R implies a second distinct pulse of the Eurasian expansion of AMH, I guess that associated with that of Y-DNA K".
Again, I've been telling you that for ages. In fact the original source of our disagreement was when I claimed R mt-hap's expansion was connected to that of K Y-hap. But once more we'd still disagree as to where that expansion originated, but we'll see where time leads us with that one too.
Thanks for all the work you've done here, and I do agree with pretty much all of what you've now concluded. But is mt-hap N derived from M or is that just a shorthand way you've described the situation?
I see you've got R's expansion at 52,000 years, at which time you say S was already in Australia. Puts the arrival in Oz at more than 52,000 years. I have no problem with that date, in fact I think it's quite likely. Megafauna extinctions would not have happened immediately. It would take time for the human population to rise to epidemic proportions.
I've been telling you that for ages, but you've only just now come to agree with me. Congratulations. Of course we'd still argue over the route the N-lineage took to reach East Asia and Australia but we'll leave that for another day.Indeed I don't believe at all your hypothesis re. N. It's just wacko and against the data.
You want to focus on N subclades because of your bias, but while those two N(xR) haplogroups were apparently already a their destinations, a wide range of M subclades were too: M3 (SA), macro-M4 (SA), M5 (SA), macro-M6 (SA), M7 (EA), M22'25 (?), M33 (SA), M35 (SA), M44'52 (?), M47'50 (?) and M51 (?).
So you have, following PhyloTree, 14 lineages that began expanding before R did, of which 11 are M-derived and only 3 N-derived (N1'5, N9 and S). Geographically you get the following distribution:
- South Asia: 4+ (most unlocated clades are surely of that extremely diverse area)
- Eastern Asia: 2
- Oceania: 1
- West/South Asia: 1 (it's pre-N1, N1 would be already West Asian, pioneering the area simultaneous to R expansion)
Later expanding N clades are also found east and west of India, so unless there was some sort of extremely bountiful oasis in Turkestan that later became absolutely deserted, there's no way to claim other origin for these lineages than South Asia, if you wish with a loose understanding of the term: the historical Eastern Indies.
Again, I've been telling you that for ages. In fact the original source of our disagreement was when I claimed R mt-hap's expansion was connected to that of K Y-hap.The source of those disagreemnts was that you claimed a Papuan origin for mtDNA R and Y-DNA K, which makes no sense. A South Asian origin and a largely joint expansion from that area make all sense instead.
But is mt-hap N derived from M or is that just a shorthand way you've described the situation?Shorthand. It just depicts mutations from L3 (all branches fused in a single unidemensional expression for convenience). I will post a proper tree when I finish with it. My assumption so far is that distance from L3 (and distance between each node) means similar time periods (that happened to be of 2.2 ky roughly, if I recall correctly).
I see you've got R's expansion at 52,000 years, at which time you say S was already in Australia. Puts the arrival in Oz at more than 52,000 years.Apaprently. Notice that it is just an estimate. Rather than assuming that all details are correct, I assume that the whole picture is close to what actually happened.
"So you have, following PhyloTree, 14 lineages that began expanding before R did, of which 11 are M-derived and only 3 N-derived".
And which of those lineages, apart from one N lineage, had reached Australia by that time? Quote: "Oceania: 1". Just because M had occupied most of mainland SE Asia doesn't at all preclude the arrival of N from the north. So how many of those N-derived haplogroups were in India?
"The source of those disagreemnts was that you claimed a Papuan origin for mtDNA R and Y-DNA K".
No I didn't. Merely a Wallacea origin. A term you were unfamiliar with at the time, in fact you thought I'd made it up.
"I will post a proper tree when I finish with it".
I'm going to send an adjusted tree to Tim today (hopefully). It will be interesting to compare them.
Regarding dating for arrival in OZ:
http://en.wikipedia.org/wiki/Australian_Aboriginal_Prehistoric_Sites
(Sorry that it's Wiki but that's the first one that came up). But here's a fairly concise look at the first Australians that you may find informative:
http://www.janesoceania.com/australia_aboriginal_origins/index1.htm
And which of those lineages, apart from one N lineage, had reached Australia by that time? Quote: "Oceania: 1".S, yah.
Just because M had occupied most of mainland SE Asia doesn't at all preclude the arrival of N from the north. So how many of those N-derived haplogroups were in India?I still do not gather how you make Australians migrate from Africa via Siberia. It's total nonsense to me. It is just totally logical that people went through South Asia (climate, shortest route, reasonable genetic evidence, total lack of evidence for your northern route idea...). Additionally, if you take the fastest expanding N subclades (N9 and S), you get SE Asia as center of gravity, not Altai.
Once and again the south appears at the center of human dispersal, what only makes total sense.
Let's put it in simple terms for dumbs (i.e. Sykes' style): Nichole had two daughters, Nora and Suzanne. We know that Nora lives in Beijing and Suzanne in Sydney and that both live at similar distance from their mother's home. Where to look for Nichole's home? I'd say somewhere around Bangkok, right?
You could argue, I guess for SE Asian urheimat for N. But this would somewhat clash with the fact that other N lineages (N1, X, W) moved westward soon after, so again the South Asian theory seems vindicated.
No I didn't. Merely a Wallacea origin. A term you were unfamiliar with at the time, in fact you thought I'd made it up. I don't see much difference, sincerely: Papua and Sulawesi are side by side. And your model is all based on ideas on a presumed high diversity of Y-DNA K in Melanesia, which we know now are wrong (all is M).
"And your model is all based on ideas on a presumed high diversity of Y-DNA K in Melanesia, which we know now are wrong (all is M)".
I presume you know very well that your comment is incorrect. It's true that M1, 2 and 3 are Papuan and Melanesian but what about K3, and S? I haven't been able to find where K2 is found. Do you have that information?
Found it:
http://genome.cshlp.org/content/18/5/830.full.pdf
The original Karafet et al reconstruction. Quote:
"Lineages K2, K3, and K4 are found in Oceania, Indonesia and/or Australia".
And S is not Australian at all. It's New Guinea, Melanesia and Indonesia. So one of the Ks (K2 or K4) is the only F-derived Y-chromosome found in Australia. Strange, isn't it.
No, I don't really know where does K2 belong to, nor K4 for the case.
Do you think that K spawned in SE Asia? That would be a reasonable hypothesis maybe. But how do you link that with your Central Asian and Siberian hypothetical routes beats me. (More at a time when CA was a Neanderthal domain, but even disregarding this).
You have some anti-Indian prejudice, no matter that some South Asians are of all peoples the ones who most resemble Australian Aborigines' appearence.
And then you have that idea that boating (not even some sort of advanced boating but the very concept of floating on something over water itself and paddling to move through it) must have been the cultural engine that somehow brought Papuans to West Eurasia. It couldn't be whatever other technology or social advantage but had to be especificall boating and the most primitive form of that art preferebly.
That I cannot swallow. I'd rather think of something more in the line of advanced guerrilla tactics like spear throwing, which at least has some evidence backing it.
"Lineages K2, K3, and K4 are found in Oceania, Indonesia and/or Australia"I see. Some may well be the Austronesian K lineages I read about long ago. At least one should be Australian.
And S is not Australian at all. It's New Guinea, Melanesia and Indonesia. So one of the Ks (K2 or K4) is the only F-derived Y-chromosome found in Australia. Strange, isn't it. Why strange?
And by the way, you seem to miss this paragraph on the ancestor of K:
Clade F
A total of 25 mutations define the lineage leading to the FT clade
(Fig. 1). Four additional subclades: F-P91/P104 (F1), F-M427/
M428 (F2), F-P96 (F3), and F-P254 (F4) were found to have the
derived state at P14 and the ancestral state at mutations that
define the G-T lineages. F-P91/P104 and F-P254 chromosomes
were found in Sri Lanka, F-M427 and M428 chromosomes were
present in East Asia (Sengupta et al. 2006), and a single individual
with the F-P96 chromosome was observed in the Netherlands.
Paragroup F* was observed primarily on the Indian subcontinent at low or moderate frequency (Kivisild et al. 2003; Karafet et al. 2005; Sengupta et al. 2006; Zerjal et al. 2007)..
How come you gather that F went via Central Asia with these results?
"I don't really know where does K2 belong to, nor K4 for the case".
K2 and K4 both from Oz/New Guinea evidently.
"But how do you link that with your Central Asian and Siberian hypothetical routes beats me".
I don't. I've never claimed "that F went via Central Asia", merely C.
"Why strange?"
Because if the Australian Y-chromosome had come via India we would expect a greater similarity between the two continents than what we find.
K2 and K4 both from Oz/New Guinea evidently.
Oceania, Indonesia and/or Australia, that's what the paper states. I recall Polynesians were the ones having high K diversity, not so much Papuans - and Polynesians are a recent phenomenon.
I don't. I've never claimed "that F went via Central Asia", merely C.I misunderstand you then: you have indeed claimed that people in general could not have reached SE Asia from South Asia because the coast between these two provinces is very hard to navigate or other pseudo-sailor arguments of the kind.
Anyhow Y-DNA C is ditributed by an area wih shape of T, between South and Eastern Asia and Sahul. Again logical thing is to realize that it spawned from SE Asia (or maybe South Asia).
Wether pre-derived C arrived to that area via Central Asia or in an UFO is something that we can't really know (a single common ancestor is enough in fact) but if you draw a line between Djibuti and Rangoon it doesnt seem to go via Altai in any case.
Because if the Australian Y-chromosome had come via India we would expect a greater similarity between the two continents than what we find. Like with all regions. Why would Australia be genetically closer than Japan or Europe? Plus there are indeed shared lineages but why do you ask more closely shared for Australia than for China? What's the point?
There is a model called "rapid coastal expansion", it's full of founder effects at the beginning of all (logically).
What I'm now thinking it may be a possibility is the inverse of your (always confusing and unfinished) model: NOP might have coalesced in Eastern Asia and P represents the first migration through the steppes into West Eurasia (and other places). That would fit well with a SE Asian origin for K but you still have L kind of misplaced and IJK causing issues at the root.
Maju said,
"What I'm now thinking it may be a possibility is the inverse of your (always confusing and unfinished) model: NOP might have coalesced in Eastern Asia and P represents the first migration through the steppes into West Eurasia (and other places). That would fit well with a SE Asian origin for K but you still have L kind of misplaced and IJK causing issues at the root."
Rejoice, for Maju has finally come to see the light! ;)
According to data published by Wen Bo and others, haplogroup F-M89(xK-M9) Y-DNA is quite widespread among Sino-Tibetan ethnic groups, including Yi (19/50 = 38.0%), Garo (6/33 = 18.2%), Northern Han (49/558 = 8.8%), Pumi (2/47 = 4.3%), Mosuo (1/47 = 2.1%), and Southern Han (11/731 = 1.5%). Haplogroup F(xK) Y-DNA has also been found among Mongols and Koreans, but in the northern parts of East Asia, at least, derived forms that are commonly seen in West Eurasia and Africa are found, e.g. J-12f2 and G-M201.
Just a possibility, that would make some sense if K is originary from SE Asia.
I don't understand why do you "rejoice" so much when you have not defended anything of the like through this thread or anywhere else that I know of. It's not like you can say "I told you": you were silent in fact.
According to data published by Wen Bo and others, haplogroup F-M89(xK-M9) Y-DNA is quite widespread among Sino-Tibetan ethnic groups, including Yi (19/50 = 38.0%), Garo (6/33 = 18.2%), Northern Han (49/558 = 8.8%), Pumi (2/47 = 4.3%), Mosuo (1/47 = 2.1%), and Southern Han (11/731 = 1.5%). Haplogroup F(xK) Y-DNA has also been found among Mongols and Koreans, but in the northern parts of East Asia, at least, derived forms that are commonly seen in West Eurasia and Africa are found, e.g. J-12f2 and G-M201. And by this you mean what?
J and G are typically thought as "silkroad" arrived, hough maybe you have some other tentative explanation.
I am more interested in the P* found among Mongols and Native Americans, which among the latter at least was P(xR1,xQ) - so either R2 or, more likely, undefined P*. Sorry but I don't keep the link to the relevant paper, though I'm sure it's listed somewhere in Ren's forum.
Maju said,
"I don't understand why do you "rejoice" so much when you have not defended anything of the like through this thread or anywhere else that I know of. It's not like you can say "I told you": you were silent in fact."
I always have opposed the faction that has tried unreasonably to force everyone to accept as "African" the origin of any Y-DNA haplogroup that has been found in any population inhabiting any territory within the boundaries that geographers have defined as "Africa." The same faction, or one which shares a similar agenda, has also tried to obscure the fact that all major subclades of haplogroup CF-M168 are represented in East Asia, but not in Africa. I suppose I should be more explicit in the future, since some people, such as Maju, have apparently failed to catch my drift.
"I recall Polynesians were the ones having high K diversity".
Possibly. But Polynesians have very little K. Mostly C2 with lesser amounts of O (although O increases as you move back along the presumed migration route). No K has been found in NZ Maori.
"you have indeed claimed that people in general could not have reached SE Asia from South Asia because the coast between these two provinces is very hard to navigate or other pseudo-sailor arguments of the kind".
It is. I've claimed that movement from South Asia to SE Asia would have been slow. This would be so whether they went via the coast, the mangrove swamps or the densely forested hill country.
"Wether pre-derived C arrived to that area via Central Asia or in an UFO is something that we can't really know".
Exactly. However an arrival via India seems unlikely to me.
"if you draw a line between Djibuti and Rangoon it doesnt seem to go via Altai in any case".
By aircraft then, obviously. Besides which the problems for migration really get difficult once you reach Rangoon, unless you have a very efficient boat.
"Why would Australia be genetically closer than Japan or Europe?"
In fact we do find that Australia is closer to Japan than it is to either India or Europe, if we look at Y-hap C anyway. Japan has a lot of O but this is probably the result of the relatively more recent expansion of O. The T-shape you claim for C is rather a sort of star-shape centred on SE Asia.
But we are getting away from mtDNA timeline, but this discussion may help you in your mission.
I always have opposed the faction that has tried unreasonably to force everyone to accept as "African" the origin of any Y-DNA haplogroup that has been found in any population inhabiting any territory within the boundaries that geographers have defined as "Africa." The same faction, or one which shares a similar agenda, has also tried to obscure the fact that all major subclades of haplogroup CF-M168 are represented in East Asia, but not in Africa. I suppose I should be more explicit in the future, since some people, such as Maju, have apparently failed to catch my drift.
I'm not here to look for subtle winks: I expect blunt explicity and I try to deliver the same. But in any case I had no idea that you espoused such hypothesis.
Guess that we will agree that BT and DE have an African oringin, right? C,F (or rather C and F, because after all hey share a single SNP) does seem to be Asian but, excepting C maybe, I see no clear hint that they would have coalesced in East Asia. Conidering how male lineages tend to drift anyhow, they could have coalesced anywhere and been drifted out in their homelands before they could consolidate.
You also have to look at mtDNA, which is even less suggestive of Eastern origins for the Eurasian fraction of humankind, though SE Asia would seem to be more the extension of South Asia in the process of expansion. I think the emphasis on South Asia can well be replaced by an emphasis on Tropical Asia or, as were known some centuries ago, the Eastern Indies or just the Indies. But I don't see any strong reason to drop South Asia from the process nor to emphasize NE Asia or Central Asia in it, being them much less diverse and in general showing the signature of more recent expansions.
It is. I've claimed that movement from South Asia to SE Asia would have been slow. This would be so whether they went via the coast, the mangrove swamps or the densely forested hill country.Was no race.
How many days would a boat take from Calcutta to Rangoon? I'm curious.
Anyhow, even if they took months, it doesn't really matter, as these migrations took time through milennia: hundreds of generations. The really important thing was that they had something to eat in the meantime and that they did not freeze in the winter.
How long would it take via the steppe, btw? Remember that via Siberia there were major fluvial obstacles and that Uyghuristan was a desert. And all that area was just freezing cold, even more than today. Only the Altai strip would seem to provide some viability for inhabitation/travel in those ages.
Exactly. However an arrival via India seems unlikely to me. However an arrival via the steppes seems impossible to me. So?
By aircraft then, obviously. Besides which the problems for migration really get difficult once you reach Rangoon, unless you have a very efficient boat.Don't be a smartass and adapt the straight line to the actual landmass. It still goes by southern Asia naturally.
I don't know what problems you mean at Rangoon other than immigration officials. The Irrawady provides there a perfectly good waterway for travel. Once saved the Ganges and Irrawady deltas it seems to me that walking might be a nice option for traveling as well, especially if you have to arrive to Tonkin before the Nth generation and they are going to find you grunting somehwere at the Kraa isthmus.
In fact we do find that Australia is closer to Japan than it is to either India or Europe, if we look at Y-hap C anyway.Erm... And if we look at F then what? And what if we look at D? And mtDNA?
You preselect your data in order to reach to your pre-cooked conclussions.
Be honest at least with yourself. You're much more likely to decieve yourself than anyone else in fact.
Maju said,
"Guess that we will agree that BT and DE have an African oringin, right?"
Yes for the former, no for the latter.
Maju said,
"You also have to look at mtDNA..."
...such as mtDNA haplogroup N9-R, one branch of which subsumes nearly all the mitochondrial diversity of Western Eurasia.
...such as mtDNA haplogroup N9-R, one branch of which subsumes nearly all the mitochondrial diversity of Western Eurasia.That if you follow Ian Logan's site but he admits that the graphs are not all updated. If you follow Phylotree, which claims to be updatd every three months, then you have no connection other than N.
I'll see to post something more detailed on mtDNA but the information available is kind of messy. There's nothing like ISOGG for mtDNA not even a logical, clear, nomenclature to follow in many cases.
At the moment I'm considering the N9-R shared mutation as a false alarm.
Yes for the former, no for the latter.
There's much more DE* in Africa than in Asia (allegedly 2 DE* in Tibet and that's all). Comparatively Africa is undersampled anyhow.
"Conidering how male lineages tend to drift anyhow, they could have coalesced anywhere and been drifted out in their homelands before they could consolidate".
And then:
"I don't see any strong reason to drop South Asia from the process nor to emphasize NE Asia or Central Asia in it, being them much less diverse and in general showing the signature of more recent expansions".
Doesn't your first comment explain how humans could have moved through Central Asia and yet leave the pattern we see today? Changing climate could well have reduced the population to just small enclaves that expanded again when good times returned.
"Don't be a smartass and adapt the straight line to the actual landmass".
It was you who first mentioned the straight line, so be prepared to suffer the consequences.
"And if we look at F then what? And what if we look at D?"
If we look at either of those Y-haps we find virtually no correspondence between India and Australia. Australia does have about a quarter of F-derived K but those are separate versions to the one found in India. And even the mtDNA shows very little correspondence. Australia is virtually all N or N-derived mtDNA with very, very few M-derived lines, while India is majority M.
"You preselect your data in order to reach to your pre-cooked conclussions".
Who do you believe is doing that? I think the evidence is pretty overwhelming that it is you who is doing that.
"At the moment I'm considering the N9-R shared mutation as a false alarm".
But R still derives somewhere from some sort of N. It's not N's sister clade. And R is very common in SE Asia. Of course that doesn't mean it originated there, but it would mean it either replaced pre-existing mtDNA lines there or, as I suspect, the region was very sparsely settled before R arrived.
Doesn't your first comment explain how humans could have moved through Central Asia and yet leave the pattern we see today?"Could" or better should have been "might" is the key word in this sentence: so extremely hypotetical, so much lacking any evidence that, yeah, it might be... but that's all. You enter the realm of no data: it might have been nearly anything at that stage.
But I prefer the solution that poses less problems: the simplest apparent answer.
It was you who first mentioned the straight line, so be prepared to suffer the consequences. Well, just got a map and traced a straight line between Djibuti and Tokyo. It only crosses one water mass: the Arabia Sea. For the rest it looks very much approximate to a realistic walk.
Then I traced a straight line from Djibuti to Arnhem peninsula in Australia. It goes mostly through the sea but again the closest approximation by land is by southern Asia.
There's no way that common sense brings you to Central Asia.
If we look at either of those Y-haps we find virtually no correspondence between India and Australia.Not true: F exists in both India and Australia/Papua, while D is absent in both regions.
And even if we look at C Australia is as similar to India as it is to Japan. I got your graph and you mark C* as SE Asian, when in fact it is South Asian (or mostly so), together with C5.
Australia does have about a quarter of F-derived K but those are separate versions to the one found in India.And to the ones found in Japan as well.
And even the mtDNA shows very little correspondence.There are lineages like N2 (macro-W), though there seem to exist also others more related to East Asia.
I have always found odd that Australia would relate to South or West Asia in ways that it does not relate to Eastern Asia, as it seems logical that all contacts between Australia and India would have to go via SE Asia. But some have left no apparent trace there, what is certainly intriguing and somehow suggestive of very rapid migrations and/or extreme drift at the early spread process.
Australia is virtually all N or N-derived mtDNA with very, very few M-derived lines, while India is majority M. Maybe but I don't follow your logic here: there are also M clades in Australia and New Guinea. West Eurasia has also almost no M and still is closest autosomally to South Asia than to aywhere else.
Who do you believe is doing that?You.
I think the evidence is pretty overwhelming that it is you who is doing that. May I ask you again to put your "evidence" in clean form and expose your model properly so the rest of us can know what exactly you're talking about and ponder what part of reason you may have and what part of delusion as well.
But R still derives somewhere from some sort of N. It's not N's sister clade. And R is very common in SE Asia.Sure, R represents a second "momentum" in the Eurasian expansion. Nevertheless this momentum also included M and N(xR) lineages: they did not stop expanding just because R did so with greater energies. I see a lot of M sublineages that either began expanding after R did or continued their expansion at that time: D, M8 (including CZ), M9(xE), E, G, Q... are just the best known ones. Much of the same can be said of N: all derived nodes but those of N9, S and N1'5 happen after the R explosion, it seems.
R rare clades (R-number) anyhow belong almost exclusively to South Asia, AFAIK. There's almost no doubt that R exploded in South Asia as South Asia holds the highest high level R diversity. I'm not fully sure but most of SE Asian R* is probably "macro-F" (F plus R9, R21 and R22) and "macro-B" (B plus R11). Do us all a favor and try to check this, please.
or, as I suspect, the region was very sparsely settled before R arrived.
It was the Paleolithic: everything was "sparsely populated". But some M and N lineages arrived before the R explosion almost for sure (rather clear in the case of N9, I'd say, but also M12'G, M9'E, M7 and possibly of M29'Q).
"I got your graph and you mark C* as SE Asian, when in fact it is South Asian (or mostly so), together with C5".
Feel free to change it. I noted C5 as being Indian because I didn't leave enough room to put S Asian.
I’ve just today noticed that it’s entirely possible to superimpose a coherent mtDNA chart over the Y-chromosome chart I sent to Tim. In other words individual mtDNA haplogroups are, in fact, often associated with individual Y-chromosome haplogroups, especially when both are expanding into previously uninhabited environments. In that situation they obviously need each other for either line to survive. The mfact the two fit together so well surely argues for some veracity in both diagrams. You will be able to use the double diagram as an aid with your project.
If you look at the diagram I sent you’ll see that all the early branches of the M group of mtDNA haplogroups are associated with Y-chromosome haplogroups enclosed within a shape a bit like a duck’s head in the middle of the diagram, Y haplogroups F*, F1, F2, F4, and most of those derived from K-T. On the other hand mtDNA haplogroup N is associated with the Y-haps at the left of the diagram and with those at the top right, and the descendant of mtR are associated with a particular haplogroup at the bottom right. I’ll try to scan a copy and send it to Tim but you can probably see the pattern for yourself.
"F exists in both India and Australia/Papua".
But in vastly different forms. And Australia and Papua are very different. You haven't looked at all closely at either of them. Drift, bottlenecks and founder effect don't really explain the difference. In other words there had been a whole lot happening there 'before the R explosion'.
I noted C5 as being Indian because I didn't leave enough room to put S Asian. Erm... check it. You wrote "SE Asian", not "Indian" nor "S. Asian".
I’ve just today noticed that it’s entirely possible to superimpose a coherent mtDNA chart over the Y-chromosome chart I sent to Tim.I give no value to that chart: it's just obviousness and some notable errors like that of C*. It tells me nothing.
In other words individual mtDNA haplogroups are, in fact, often associated with individual Y-chromosome haplogroups, especially when both are expanding into previously uninhabited environments.I wish it was so easy.
You will be able to use the double diagram as an aid with your project.I have tried before. It is not that easy, really.
If you look at the diagram I sent you’ll see that all the early branches of the M group of mtDNA... It is a Y-DNA diagram. There's no mtDNA M in it.
...enclosed within a shape a bit like a duck’s head...uh?
Y haplogroups F*, F1, F2, F4, and most of those derived from K-T. On the other hand mtDNA haplogroup N is associated with the Y-haps at the left of the diagram and with those at the top right, and the descendant of mtR are associated with a particular haplogroup at the bottom right. I’ll try to scan a copy and send it to Tim but you can probably see the pattern for yourself.No I see nothing of what you say.
Also it'd be nice if you drawed instead of using Word to "draw". If you don't like Paint, I can send you a pirated copy of Autocad2002, which is ideal for detailed design with infinite zoom/unzoom and even tri-dimensional possibilities, or you can use some drawing freeware like Gimp.
"F exists in both India and Australia/Papua".
But in vastly different forms.Yah, exactly like C or nearly any other clade.
And Australia and Papua are very different.But they share more among them than with any region outside, maybe excepted some Negrito remnants in the Malay archipielago. Whatever the case both represent the signature of early Eurasians flowing through SE Asia in southward direction. You can't pick one and ignore the other at your convenience. Always consider both, please.
Drift, bottlenecks and founder effect don't really explain the difference.Founder effects can. We are after all talking of small groups migrating into a vast empty region. That will unavoidably cause a founder effect or a series of them.
Correction:
I said:
''I noted C5 as being Indian because I didn't leave enough room to put S Asian.
Erm... check it. You wrote "SE Asian", not "Indian" nor "S. Asian".''
I was thinking of C* sorry. My caveat is that C* is South Asian (Indian if you wish), not SE Asian. With such essential errors you get to essentially wrong conclussions.
Maju said,
"I was thinking of C* sorry. My caveat is that C* is South Asian (Indian if you wish), not SE Asian. With such essential errors you get to essentially wrong conclussions."
This is factually incorrect; the frequency of haplogroup C(xC1, C2, C3, C4, C5) is much higher in Southeast Asia, including southern China, than it is in India. I can provide relevant data if you wish.
I can provide relevant data if you wish.
For sure. I was only aware of C* in India, not in SE Asia.
"it's just obviousness and some notable errors like that of C*".
Hardly a notable error. Thanks for your comment regarding C* Ebizur. And regarding some relevant figures for Australia you provided some time ago:
Y-hap C at 61.7% and K (presumably K2 and K4) at 27%. That makes nearly 90%.
Minority clades include R at 8.5%, F(xK) at just 2.1% and even O at 0.7%. These last three could easily be more recent arrivals with the development of trade routes around S and SE Asia, including those operated by Europeans. The ever practical Dutch, especially, tended to use local sailors in their trading vessels.
Because the Australian K Y-haps are downstream mutations within their hapologroups and they have not become significantly differentiated within the region they too presumably arrived after Y-hap C. And it's interesting that K is not found in New Guinea. Suggests it didn't arrive in Australia by that route.
So where's the connection with India Maju? Your comment: "F exists in both India and Australia/Papua" is incorrect.
"check it. You wrote 'SE Asian', not 'Indian' nor 'S. Asian'".
Check again Maju. For C5 I wrote 'India'. It was C* where I wrote 'SE Asia'. Who's not checking their facts now? I've since added 'S Asia' to C*, just for your benefit.
"It is a Y-DNA diagram. There's no mtDNA M in it".
There is now. Have a look.
"You can't pick one and ignore the other at your convenience. Always consider both, please".
The evidence is pretty overwhelming that you consistently completely ignore both.
"I was only aware of C* in India, not in SE Asia".
A perfect example of your ignoring the region. As I've said many times, you can prove anything if you ignore Australia. I should add New Guinea and SE Asia to my saying. You're only interested in what's happened in Europe. Much had happened elsewhere by the time modern haplogroups reached that region.
Regarding the similarity, or otherwise, of Australia and Melanesia:
http://www.cell.com/AJHG/abstract/S0002-9297(07)62482-3
Maju said,
"For sure. I was only aware of C* in India, not in SE Asia."
Actually, it seems that haplogroup C* is hardly found in South Asia. Here is what Sengupta et al. (2006), who have first reported haplogroup C5-M356, have to say about it:
"Discovery of Five New Clades That Improve HG Topology
We report five new clades that improve the HG topology within the Y-chromosome genealogy. The new subclade C5-M356 accounts for 85% of the former C* HGs. Although its overall frequency is only 1.4% in the Indian sample, it occurs in all linguistic groups and in both tribes and castes. It also occurs in one Dravidian Brahui in Pakistan (table 5)."
Note that the "1.4%" here is an error for 1.51%; this paper is actually full of errors, which makes me hesitant to rely upon it, but, sadly, nearly all sources on the Y-DNA of South Asia are equally unreliable."C5-M356 Is an Ancient HG That Originated in India
The phylogeography, frequency (tables 5 and 6), and age of Y-microsatellite variation (table 11) of HG C5-M356 lineages are indicative of an in situ Indian origin and considerable antiquity. HG C(xC3) lineages occur at considerable frequency in Australia (Kayser et al. 2000). Some have interpreted that the paraphyletic HG C(xC3) reflects the Paleolithic colonization event concerning the peopling of Australia via a southern coastal migration route involving India (Kivisild et al. 1999, 2003a; Underhill et al. 2001a), whereas others (Redd et al. 2002) argued that such C*-defined chromosomes reflects recent (≤5 KYA) genetic affinity between Indian tribes and Australian Aborigines. Interestingly, the Australian Aborigines with the C(xC3) HG display a large multistep deletion at the DYS390 microsatellite locus (Forster et al. 1998) that is considered to be a unique mutational event (Kayser et al. 2000), whereas the Indians do not. These Australian DYS390.1-deletion HG C chromosomes lack the M356 mutation (M. Kayser, personal communication), which undermines claims of a recent common shared ancestry with India. Conversely, the C5 data support a more-ancient affinity that is consistent with that observed with respect to mtDNA polymorphisms (Macaulay et al. Macaulay et al., 2005 V Macaulay, C Hill, A Achilli, C Rengo, D Clarke, W Meehan, J Blackburn, O Semino, R Scozzari, F Cruciani, A Taha, NK Shaari, JM Raja, P Ismail, Z Zainuddin, W Goodwin, D Bulbeck, HJ Bandelt, S Oppenheimer, A Torroni and M Richards, Single rapid coastal settlement of Asia revealed by analysis of complete mitochondrial genomes, Science 308 (2005), pp. 1034–1036. View Record in Scopus | Cited By in Scopus (119)2005)."
Chamar (India, North; Low caste; Indo-European)
1/18 = 5.6% C5-M356
8/18 = 44.4% H1-M52
7/18 = 38.9% R1a1a-M17
2/18 = 11.1% R2-M124
Rajput (India, North; High caste; Indo-European)
1/29 = 3.4% C5-M356
1/29 = 3.4% F*-M89/M213
1/29 = 3.4% H-M69
5/29 = 17.2% H1-M52
1/29 = 3.4% J1-M267
1/29 = 3.4% J2a-M410
3/29 = 10.3% J2b2-M241
2/29 = 6.9% L3-M357
1/29 = 3.4% O3a3c-M134
9/29 = 31.0% R1a1a-M17
4/29 = 13.8% R2-M124
Chakma (India, Northeast; Tribe; Tibeto-Burman)
1/4 = 25% C5-M356
3/4 = 75% O3a3c-M134
Lodha (India, East; Tribe; Austro-Asiatic)
1/20 = 5.0% C5-M356
2/20 = 10.0% H-M69
3/20 = 15.0% H1-M52
7/20 = 35.0% J2b2-M241
7/20 = 35.0% R2-M124
Mahishya (India, East; Middle caste; Indo-European)
1/13 = 7.7% C5-M356
1/13 = 7.7% H-M69
6/13 = 46.2% H1-M52
1/13 = 7.7% J2b2-M241
1/13 = 7.7% O2a-M95
1/13 = 7.7% R1a1a-M17
2/13 = 15.4% R2-M124
Irula (India, South; Tribe; Dravidian)
1/30 = 3.3% C5-M356
9/30 = 30.0% F*-M89/M213
8/30 = 26.7% H1-M52
5/30 = 16.7% H2-APT
1/30 = 3.3% J1-M267
3/30 = 10.0% L1-M76
3/30 = 10.0% R2-M124
Iyer (India, South; High caste; Dravidian)
2/29 = 6.9% C5-M356
3/29 = 10.3% G2a-P15
1/29 = 3.4% H-M69
1/29 = 3.4% H1-M52
4/29 = 13.8% J2a-M410
1/29 = 3.4% J2b2-M241
5/29 = 17.2% L1-M76
1/29 = 3.4% R*-M207
8/29 = 27.6% R1a1a-M17
3/29 = 10.3% R2-M124
Pallan (India, South; Low caste; Dravidian)
1/29 = 3.4% C*-M216/RPS4Y
2/29 = 6.9% F*-M89/M213
4/29 = 13.8% H1-M52
1/29 = 3.4% H2-APT
1/29 = 3.4% J2a-M410
3/29 = 10.3% J2b2-M241
4/29 = 13.8% L1-M76
1/29 = 3.4% L3-M357
7/29 = 24.1% R1a1a-M17
1/29 = 3.4% R1b1b2-M269
4/29 = 13.8% R2-M124
Toda (India, South; Tribe; Dravidian)
1/8 = 12.5% C*-M216/RPS4Y
2/8 = 25.0% J2a-M410
4/8 = 50.0% L1-M76
1/8 = 12.5% R1a1a-M17
Vanniyar (India, South; Middle caste; Dravidian)
1/25 = 4.0% C5-M356
4/25 = 16.0% F*-M89/M213
2/25 = 8.0% H-M69
4/25 = 16.0% H1-M52
2/25 = 8.0% J2a-M410
2/25 = 8.0% J2b2-M241
5/25 = 20.0% L1-M76
2/25 = 8.0% R1a1a-M17
3/25 = 12.0% R2-M124
Maratha (India, West; Middle caste; Indo-European)
1/20 = 5.0% C5-M356
1/20 = 5.0% F*-M89/M213
5/20 = 25.0% H1-M52
1/20 = 5.0% H2-APT
1/20 = 5.0% J2a-M410
2/20 = 10.0% J2b2-M241
3/20 = 15.0% L1-M76
2/20 = 10.0% R1a1a-M17
4/20 = 20.0% R2-M124
Nav Buddha (India, West; Low caste; Indo-European)
1/14 = 7.1% C5-M356
1/14 = 7.1% G2a-P15
6/14 = 42.9% H1-M52
1/14 = 7.1% J2a1d-M158
1/14 = 7.1% J2b2-M241
1/14 = 7.1% L1-M76
3/14 = 21.4% R1a1a-M17
Brahui (Pakistan, South; Sengupta et al. 2006):
1/25 = 4.0% C5-M356
1/25 = 4.0% G1-M285
3/25 = 12.0% G2a-P15
1/25 = 4.0% H1*-M52
3/25 = 12.0% J1-M267
2/25 = 8.0% J2a-M410
2/25 = 8.0% J2a1b-M67
1/25 = 4.0% L1-M76
1/25 = 4.0% L3-M357
1/25 = 4.0% O3-M122
6/25 = 24.0% R1a1-M17
3/25 = 12.0% R2-M124
Burusho (language isolate)
Sengupta et al. (2006):
1/20 = 5% C3-M217
1/20 = 5% G2c-M377
2/20 = 10% H1*-M52
1/20 = 5% H1a1-M197
1/20 = 5% J2b2-M241
1/20 = 5% K*-M9
3/20 = 15% L3-M357
1/20 = 5% O3a3c-M134
2/20 = 10% R1a1-M17
3/20 = 15% R2-M124
4/20 = 20% R*-M207
Hazara (Indo-Iranian > Iranic > Western Iranic > SW Iranic)
Sengupta et al. (2006):
10/25 = 40% C3-M217
1/25 = 4% I2b1b-M379
1/25 = 4% J2a-M410
2/25 = 8% O3-M122
1/25 = 4% Q1a1-M120
1/25 = 4% Q1b-M378
8/25 = 32% R1b1b1-M73
1/25 = 4% R2-M124
Pathan (Indo-Iranian > Iranic > Eastern Iranic > SE Iranic)
Sengupta et al. (2006):
1/20 = 5% C3-M217
1/20 = 5% G2a-P15
1/20 = 5% G2c-M377
1/20 = 5% H1*-M52
1/20 = 5% H*-M69
1/20 = 5% L1-M76
1/20 = 5% L3-M357
2/20 = 10% Q1a3-M346
8/20 = 40% R1a1-M17
2/20 = 10% R1b1b2-M269
1/20 = 5% R*-M207
India total:
11/728 = 1.51% C5-M356
2/728 = 0.27% C-M216/RPS4Y(xC1-M8, C3-M217, C5-M356)
Pakistan total:
12/176 = 6.82% C3-M217(xC3a-M93, C3d-M407)
Haplogroup C* has been found in this study only in one Pallan and one Toda, both of which are Dravidian-speaking populations of South India; the Pallans are a rather low-ranking caste of cultivators from Tamil Nadu (I have read that the Pallans are considered to be the "highest among the low castes" in Tamil Nadu), whereas the Todas are a buffalo-herding pastoralist tribe from the Nilgiri Hills, which are located around the junction of the three South Indian states of Tamil Nadu, Karnataka, and Kerala.
Sharma et al. (2009) have not reported finding any C(xC5) among their samples of Indian populations; however, this study is also plagued by the same sorts of errors as the former study by Sengupta et al. In any case, the data sets of Sengupta et al. (2006) and Sharma et al. (2009) suggest that 85% to nearly 100% of Indian haplogroup C Y-DNA belongs to the subclade C5-M356. Nearly all the haplogroup C Y-DNA in Pakistan, on the other hand, seems to belong to C3-M217(xC3a-M93, C3c-M48, C3d-M407) according to the data of Sengupta et al. (2006) and Qamar et al. (2002).
Tenzin Gayden et al. (2007) have reported finding haplogroup C-M216(xC3-M217, C5-M356) in 3/77 = 3.9% of a sample of people from Kathmandu, Nepal, but this amounts to a total of only 3/188 = 1.6% C(xC3, C5) for the pool of their three samples from Nepal:
Tamang:
2/45 = 4.4% F-M213(xH-M69, J-M304, K-M9)
1/45 = 2.2% O3a3c-M134(xO3a3c1-M117)
38/45 = 84.4% O3a3c1-M117
2/45 = 4.4% R2-M124
2/45 = 4.4% R1a1-M198
Newar:
2/66 = 3.0% C5-M356
4/66 = 6.1% H1a-M82
4/66 = 6.1% J2a-M410
1/66 = 1.5% J2b2-M241
14/66 = 21.2% O3a3c1-M117
17/66 = 25.8% R2-M124
17/66 = 25.8% R1a1-M198
7/66 = 10.6% R1b1b2-M269
Kathmandu:
3/77 = 3.9% C-M216(xC3-M217, C5-M356)
2/77 = 2.6% C3-M217
1/77 = 1.3% C5-M356
1/77 = 1.3% H-M69(xH1-M52)
4/77 = 5.2% H1-M52(xH1a-M82)
4/77 = 5.2% H1a-M82
3/77 = 3.9% J2a-M410
5/77 = 6.5% J2b2-M241
1/77 = 1.3% N-M231(xN1c-Tat)
1/77 = 1.3% N1c1a-P21
1/77 = 1.3% O3a-M324(xO3a3c-M134)
2/77 = 2.6% O3a3c-M134(xO3a3c1-M117)
13/77 = 16.9% O3a3c1-M117
1/77 = 1.3% Q-M242(xQ1a1-M120, Q1a3-M346)
8/77 = 10.4% R2-M124
27/77 = 35.1% R1a1-M198
Sengupta et al. (2006) have reported finding only one example of haplogroup C(xC1, C3, C5) among their East Asian samples, namely in their sample of Dai people, a group of Tai-speaking populations who live in the southwestern corner of Yunnan Province, near the borders of Burma and Laos:
Dai (Kradai > Tai)
3/7 = 42.9% O2a-M95
1/7 = 14.3% O2*-M268(xO2a-M95, O2b-M176)
1/7 = 14.3% O3-M122(xO3a3c-M134)
1/7 = 14.3% C3-M217(xC3a-M93, C3d-M407)
1/7 = 14.3% C*-M216/RPS4Y
However, Hammer et al. (2005) have reported the following examples of C-RPS4Y(xC1-M8, C2-M38, C3-M217) in their samples from China and Southeast Asia:
Tujia (Sino-Tibetan)
3/49 = 6.1% C(xC1, C2, C3)
Miao (Hmong-Mien > Hmongic)
1/58 = 1.7% C(xC1, C2, C3)
Yao (Hmong-Mien > Mienic)
12/60 = 20.0% C(xC1, C2, C3)
Malay (Austronesian > Malayo-Polynesian)
1/32 = 3.1% C(xC1, C2, C3)
Indonesia, East (Austronesian > Malayo-Polynesian)
3/55 = 5.5% C(xC1, C2, C3)
Indonesia, West (Austronesian > Malayo-Polynesian)
1/25 = 4.0% C(xC1, C2, C3)
"Southeast Asia" (including southern China) total:
21/683 = 3.1% C(xC1, C2, C3) [including Yao sample]
9/623 = 1.4% C(xC1, C2, C3) [excluding Yao sample]
Uygur (Turkic > Southeastern/Uzbek-Uyghur)
1/67 = 1.5% C(xC1, C2, C3)
I will add the data from Xue et al. (2006) later if you remain unconvinced, but haplogroup C* seems to be more common in Southeast Asia than in South Asia, although it is very infrequent in both regions. Within South Asia, haplogroup C* seems to be limited to occurring with very low frequency among some populations in northeasterly areas (such as the general population of Kathmandu, Nepal) and among some Dravidian populations in the vicinity of Tamil Nadu. The frequency of haplogroup C* among the Nepalese might rival the frequency of this haplogroup among Southeast Asians; however, it seems to be practically absent elsewhere in South Asia.
Thanks a lot, Ebizur. Your capability to keep track of all that scattered data never stops impressing me.
I take good note.
@Terry:
Regarding the similarity, or otherwise, of Australia and Melanesia:
http://www.cell.com/AJHG/abstract/S0002-9297(07)62482-3.
Yah, we do know that Australia and Melanesia are rather different. Still the continent was a single landamass until not long ago and some clades are clearly shared. The timeline of colonization of both provinces also appears to have been of similar time and age, independently if they arrived by a single or two different routes.
And regarding some relevant figures for Australia you provided some time ago:
Y-hap C at 61.7% and K (presumably K2 and K4) at 27%. That makes nearly 90%.
Minority clades include R at 8.5%, F(xK) at just 2.1% and even O at 0.7%. These last three could easily be more recent arrivals with the development of trade routes around S and SE Asia, including those operated by Europeans. The ever practical Dutch, especially, tended to use local sailors in their trading vessels.
Well, doubt their native sailors were Australian Aborigines. Much more likely they were Malays, who had a much better grasp of sailing. I still don't see how you gather that local men employed by Europeans would have been responsible of transmiting European Y-DNA.
Don't know the exact source but the McDonalds map reports that Australian Aboriginal R is R(xR1), what makes it virtually impossible to be European by origin. It still could be Indian R2 or some other unknown R lineage (that would be interesting to know about certainly). That non-Europeanness of Australian R also makes more likely that its F is non-European as well.
Australian K is mostly K4, as far as I know, if not exclusively.
And it's interesting that K is not found in New Guinea.It is: M, S and K3. In fact Melanesia is almost only Y-DNA K!
So where's the connection with India Maju? Your comment: "F exists in both India and Australia/Papua" is incorrect. No it is not: K is F, can you understand at least that? F has several clades, AFAIK: F1, F2, F3, F4, G, H and IJK. Saying "K" is like saying "F7b". And K subclades could well be called (just a silly excercise but seems you need it): F7b1 (K1), F7b2 (K2), F7b3 (K3), F7b4 (K4), F7b5 (L), F7b6 (M), F7b7 (NOP), F7b8 (S) and F7b9 (T). Keeping this in mind (while unnecesarily complex for writing down and certainly unconventional) helps a lot when comparing with other similarly old lineages such as C, D and E.
Check again Maju. For C5 I wrote 'India'. It was C* where I wrote 'SE Asia'. Who's not checking their facts now? I've since added 'S Asia' to C*, just for your benefit.Ok. Acknowledged.
I have forgotten to mention that, although Sengupta et al. have found haplogroup C* in only one Dai individual among all their East Asian samples, this still represents a greater fraction of the total sample than the two Indian C* individuals (1/175 = 0.57% C* for the pool of East Asian samples; 2/728 = 0.27% C* for the pool of Indian samples).
Also, I suppose I should point out the fact that the C(xC1, C2, C3) individuals in the Southeast Asian data set of Hammer et al. (2005) have not been tested for M356. I hope that researchers will test this SNP on samples of haplogroup C from Southeast Asia, Central Asia, and the Mediterranean (Lebanon, Turkey, Greece, Sicily, Sardinia, etc.) in the future.
Anyway, here are the data from Xue et al. (2006):
C*-RPS4Y(xC1-M8, C3-M217)
Daur: 0/39
Ewenki: 1/26 = 3.8% (This is a sample of Evenks from northeastern Inner Mongolia, probably the Hulunbuir region.)
Hezhe: 3/45 = 6.7% (This is a sample of Nanais from the northeasternmost corner of the PRC.)
Hui: 4/35 = 11.4% (This is a sample of Chinese Muslims, probably from the Ningxia Hui Autonomous Region.)
Manchu: 0/35
Inner Mongolian: 0/45
Oroqen: 0/31
Uygur (Urumqi): 1/31 = 3.2%
Uygur (Yili): 1/39 = 2.6%
Xibe: 0/41
Han (Harbin): 0/35
Han (Yili): 0/32
Korean (China): 0/25
Buyi: 0/35
Hani: 0/34
Li: 0/34
Qiang: 0/33
She: 0/34
Tibetans: 0/35
Yao (Bama): 6/35 = 17.1% (This is a sample of Yao people from Bama Yao Autonomous County in Guangxi.)
Yao (Liannan): 0/35 (This is a sample of Yao people from Liannan Yao Autonomous County in Guangdong.)
Han (Chengdu): 0/34
Han (Lanzhou): 0/30
Han (Meixian): 0/35
Japanese: 0/47
Korean (South Korea): 0/43
Outer Mongolian: 0/65
Total: 16/988 = 1.62%
However, these samples also have not been tested for C5-M356 or other known subclades of haplogroup C, so it is not really proper to call them C*.
The Yaos of Bama, Guangxi also display some other genetic peculiarities, such as a very high frequency of haplogroup O2*-P31(xO2a-M95, O2b-SRY465): 13/35 = 37.1%. They also have a high frequency of haplogroup O3a3b-M7 (12/35 = 34.3%), which is typically found in Hmong-Mien/Miao-Yao populations.
For comparison, the sample of Yaos from Liannan, Guangdong shares the high frequency of haplogroup O3a3b-M7 (18/35 = 51.4%), but completely lacks haplogroups C(xC1, C3) and O2*(xO2a, O2b). The Yao sample of Hammer et al. (2005) includes equal proportions of C(xC1, C2, C3) and O2*(xO2a, O2b): 12/60 = 20.0%.
Thanks again, Ebizur. It is indeed true that not all clades, including "Indian" C5 have been looked for and that the whole C5 tree seems unresolved to some extent. I am nevertheless inclined to consider that C, like D and NO, had a SE Asian urheimat: at least its distribution could suggest that.
Erratum:
... the whole C5 tree seems unresolved...
should read:
... the whole C tree seems unresolvedOops!
Maju said,
"Thanks again, Ebizur. It is indeed true that not all clades, including "Indian" C5 have been looked for and that the whole C tree seems unresolved to some extent. I am nevertheless inclined to consider that C, like D and NO, had a SE Asian urheimat: at least its distribution could suggest that."
Yes, I agree that more work needs to be done to elucidate the development of the phylogeny below C-M130. Testing for all presently known subclades of haplogroup C would be a good start, and then someone should try to find some new SNPs to subdivide any remaining C*, and especially try to find some SNPs to group known subclades together into intermediate branches, so that, for example, haplogroup C1-M8 and haplogroup C3-M217 might become haplogroup C1a-M8 and C1b-M217 if they are found to share an SNP to the exclusion of other subclades of C-M130, and so forth.
Actually, while I was reviewing the data of Xue et alii (2006), I noticed that most of their examples of haplogroup C(xC1, C3) have come from populations of northern China (Evenks, Nanais, Huis, Uyghurs) rather than the south, but the high frequency of C(xC1, C3) among the Yao in Bama, Guangxi is still curious, and the data of Hammer et al. (2005) and Sengupta et al. (2006) do suggest that haplogroup C* should be more common in Southeast Asia than in India. The high diversity of haplogroup C in Nepal, including C3-M217, C5-M356, and C(xC3, C5), is also interesting, but it might be due to converging streams of migration originating from several different sources and introducing already differentiated subclades of haplogroup C into the sub-Himalayan region. I think Indonesia also contains at least three different subclades of haplogroup C: C2-M38, C3-M217, and C(xC2, C3).
By the way, I would say that the major peculiarity of the Indian Y-DNA pool is not haplogroup C5-M356 nor any other sort of haplogroup C, but rather the high frequency in the subcontinent of haplogroup F(xG, I, J, K), including H-M69 and F(xG, H, I, J, K), which is hardly found anywhere else in the world.
By the way, I would say that the major peculiarity of the Indian Y-DNA pool is not haplogroup C5-M356 nor any other sort of haplogroup C, but rather the high frequency in the subcontinent of haplogroup F(xG, I, J, K), including H-M69 and F(xG, H, I, J, K), which is hardly found anywhere else in the world.
Actually that sounds a bit like a self-fulfilling prophecy because the clades you excluded from F in the enunciation are precisely those that are found outside South Asia.
But you are on something because South Asia does indeed include also most of what you excluded: at least J and K (L, O, R1, R2, Q). Though it is thought that most of that would be product of back-migrations, in fact they must have been pretty old "back-migrations".
I was recently discussing the case of J2b (and some J2a as well) in regards to South Asia at Mathilda's and if they represent a back-migration from West Asia it may be a very old one.
R2 is clearly South Asian and probably all R to begin with (R1a has also been claimed to be older in South Asia, even if the most common sublineage would be European/Central Asian, plus I understand that there is enough unresolved R* and R1* in the region to consider R as likely South Asian by origin).
Not very sure about Q because I have not dug much into this clade but from memory there were some who argued that it was too diverse to be just a back-migration issue. If so all P could be from that origin.
Even South Asian O raises some eyebrows though in general is accepted as an Austroasiatic "recent" arrival. Wonder if there is some N or NO* and even M or pre-M, so we would have the whole F macro-haplogroup represented in South Asia. Probably not but still it's close.
Anyhow I am indeed interested in digging on the F* issue. Hope some news appear before all reserach budgets are just axed on grounds of economic constraints.
"Haplogroup C* has been found in this study only in one Pallan and one Toda, both of which are Dravidian-speaking populations of South India".
Rather argues for movement into India from SE Asia, rather than the other way round.
"Still the continent was a single landamass until not long ago".
I'm beginning to doubt that. That diagram of the combined mtDNA and Y-haps I've been mucking around with at remotecentral reveals quite deep differences. It looks as though only Y-hap K and mtDNA P arrived in both places, and that probably after both places had already been settled.
Sure, the Arafura Sea is fairly shallow but at times of lower sea level it was probably more like what Southern New Guinea and the Gulf of Carpentaria are today rather than being dry land, i.e. mangrove swamp.
"Well, doubt their native sailors were Australian Aborigines".
You're not really thats stupid are you? Obviously if the sailors were Aborigines they'd hardly be likely to introduce any new Y-chromosomes to Australia.
"Much more likely they were Malays".
Exactly. Hence the Y-haps O and F(xK). But they weren't all Malays. Many were Europeans. Hence the Y-hap R. Of course any of the haplogroups could have come from elsewhere, but Unlikely to be ancient Australian.
"In fact Melanesia is almost only Y-DNA K".
Unlike Australia. Actually I did realise after I'd posted that comment that it wasn't strictly correct. I'm not sure that Y-hap S is actually found in Australia though but you claim (probably correctly) that Australia has only K4 and New Guinea has K3. Admittedly that haplogroup is relatively downstream. Similar?
"F7b1 (K1), F7b2 (K2), F7b3 (K3), F7b4 (K4), F7b5 (L), F7b6 (M), F7b7 (NOP), F7b8 (S) and F7b9 (T)".
Ok, so which of these are shared between India, Australia and New Guinea? Perhaps I should have said, 'India and Australia don't share any derived haplogroups within the F clade'.
"But you are on something because South Asia does indeed include also most of what you excluded".
Doesn't that again indicate migration into India has been relatively common rather than indicating India is a source of haplogroup variation?
Hence the Y-haps O and F(xK).
O is surely Austronesian but Austronesians have been sailing those waters since long before the whiteman arrived. I see no reason to appeal to European presence to explain haplogroups from a neighbouring area full of excellent transoceanic sailors.
Many were Europeans. Hence the Y-hap R.
Well, it seems that the R is R(xR1) - I wrote that above, so you should be aware by now - so it's not likely to be European at all. If it's R2 it should be Indian: think Vijayanagar maybe, whose merchants traded all through the Indian ocean, or maybe something else (R3?). R(xR1), including R2, is a poorly stuied clade, yet an intriguing one. Notice that R(xR1) was also found at low ammounts among NW Amercian Natives - and there was yet another author who connected it via haplotypes to Mongolian P*, though I'm not sure how valid is this.
"In fact Melanesia is almost only Y-DNA K".
Unlike Australia.
There is a lot of K in Australia too (25% in our favorite reference map).
And I will retrieve my claim that is K4: I really do not know (probably confused with C4 in fact). Karafet says that: "Lineages K2, K3, and K4 are found in Oceania, Indonesia, and/or Australia". It's all I actually know and a good reason to think of Sahul as a single continent, even if with two regions, and related to Indonesia/Polynesia.
Ok, so which of these are shared between India, Australia and New Guinea? Perhaps I should have said, 'India and Australia don't share any derived haplogroups within the F clade'.
Exactly. Because if you are building a theory on C or D as whole (not on C1, C2, C3, etc.), you will have to be consistent and also use F as a whole and not its various subclades.
East Asia and Australia do not share any derived haplogroups within the C clade. They only share root-level C, the same that they share root-level F.
Anyhow we still have the issue of F* and R*, which I think it's far from being satisfactorily solved.
In the mtDNA side of things, instead, Australia does share some derived lineages with South Asia and West Eurasia (but does not seem to share them with East Asia, curiously enough). These are:
· M14 (Australia and Arabia)
· N2 (N2a in Australia, W in South Asia and West Eurasia)
· R31 (in South Asia and Australia)
Notice that all mtDNA connections are with Australia, not New Guinea. They are minor clades indeed but there are other items (like some phenotype elements and even blood group A) that may suggest a stronger than normally assumed connection between Australia and South/West Asia.
Add to that the mysterious Y-DNA R(xR1) and you have a nice puzzle to chew on.
Doesn't that again indicate migration into India has been relatively common rather than indicating India is a source of haplogroup variation? .
Depends on how you read it. Until recently it was assumed that all R1a coalesced in Eastern Europe and expanded only with IEs. But recently it's been claimed that there is greater top level haplotype divesity within this clade in certain groups of India, so it could well mean that even if IEs coalesced at Samara and caried maybe R1a1a around the world they were ultimately original from India anyhow, probably at a Paleolithic stage.
Most shared lineages of South Asia have been often happily assumed as external inputs. This may be true in some cases (O, J, R1a1a) but others are anything but clear in fact.
"I see no reason to appeal to European presence to explain haplogroups from a neighbouring area full of excellent transoceanic sailors".
You've missed the point. I'm saying that those haplogroups were not amoung the early ones across Wallace's Line. They've arrived later.
"There is a lot of K in Australia too".
In fact that is 'unlike Australia'. 25% is nowhere near the level in New Guinea. Besides which it's becoming obvious to me that K arrived in the region after Australia was first settled, possibly long after.
"you will have to be consistent and also use F as a whole and not its various subclades".
That's would be completely unproductive for sorting out settlement patterns. Those 'various subclades' are nearly all specific to particular geographic regions. Looking at their distribution can reveal a great deal about ancient human migration. You must have noticed that fact for mtDNA as you assembled the tree.
"Australia does share some derived lineages with South Asia and West Eurasia".
But the surprising thing is that those connections are far from common. They too could be relatively recent arrivals. And as you say, 'Notice that all mtDNA connections are with Australia, not New Guinea'. Now that's interesting isn't it? How do you explain it?
I've finally worked out what it is that you cannot see. At remotecentral you wrote:
"put some beans of several colors in a bag and blindly pick some of them, discarding the rest. Then repeat with the remainder (the survivors) as many times as you want".
You used this example in order to claim that lots of haplogroups moved together, and drift then operated. But your example of the beans describes founder effect perfectly, rather than describing genetic drift. And founder effect is exactly what we would expect to find at any expanding margin of ancient human settlement. By the time any human group was able to expand into a new environment the group could well be reduced to just one or two haplogroups. Founder effect would also operate in any human group with new technology able to expand through regions already occupied. But that expansion is more likely to involve several haplogroups.
My diagram of the combined YDNA and mtDNA actually allows us to extract much of that information. And I'll send Tim a revised one in the next few days. Originally I made the concession of fitting y-hap C to a southern route but when I redesigned it for a northern route something very interesting happened with the mtDNA. Between us we will even be able to work out the pattern of migrations across Wallace's Line and into Australia, New Guinea, Melanesia and the wider Pacific. That will help us interpret the information for other regions.
You've missed the point.
I did not miss anything: you were very explicit about Ducth sailors.
Besides which it's becoming obvious to me that K arrived in the region after Australia was first settled, possibly long after. You must have some info I lack. Because it is not "obvious" to me.
"you will have to be consistent and also use F as a whole and not its various subclades".
That's would be completely unproductive for sorting out settlement patterns. Those 'various subclades' are nearly all specific to particular geographic regions.C subclades (and even D subclades to an extent) are too.
Looking at their distribution can reveal a great deal about ancient human migration. You must have noticed that fact for mtDNA as you assembled the tree.But always in parallel: M with N, R with M and the other N subclades, etc. Otehrwise you'd be comparing apples and oranges.
"Australia does share some derived lineages with South Asia and West Eurasia".
But the surprising thing is that those connections are far from common.No. The surprising thing is that in fact they are too common: not one but three clades, one of them sister of W. There's nothing of the like re. East Asia.
'Notice that all mtDNA connections are with Australia, not New Guinea'. Now that's interesting isn't it? How do you explain it?I don't have a good explanation atm. I just state the facts.
... your example of the beans describes founder effect perfectly, rather than describing genetic drift. Founder effect would be if all (or most of) the beans in the bag would be the same color to begin with.
I think my example explains drift pretty well.
And founder effect is exactly what we would expect to find at any expanding margin of ancient human settlement.Maybe. But when a population has so many lineages as Australians (they do at least mtDNA-wise), founder effect alone does not seem a really good explanation. At least on first sight: it would mean many founder effects and, considering the difficulties of crossing into Australia (or New Guinea for the case), I am not naturally inclined to consider many settler journeys, rather few instead.
My diagram of the combined YDNA and mtDNA actually allows us to extract much of that information.Not useful for me. Moreso when I don't agree with your arbitrary combinations.
Between us we will even be able to work out the pattern of migrations...I am actually doing it mostly alone. Of course I am not so much focused in the map as I am in the patterns and I think I have finally some found parallel between Y-DNA and mtDNA expansions, at least in the temporal frame.
Points of liely coincidence:
1. Y-DNA B-CF'DE split parallels mtDNA L2-L3'4 divison.
2. Y-DNA D-E split parallels mtDNA L3 branching.
3. Y-DNA D and F (and IJK) splits parallels mtDNA M (and N) explosion.
4. Y-DNA K (and NOP) explosion parallels the mtDNA R one.
These four "coupled" nodes appear to fit very well when you superimpose one tree to the other. but this "coupling" pose some other challenges:
· C and F must have been lurking in the background well before the OOA and must have been drifted out (by E, I guess) in Africa later on. The OOA did not happen therefore (by this model) at the CF'DE node but at the DE one, with C and F being minor clades that somehow got lucky with the founder effect thingy.
· Derived Y-DNA lineages that I had managed to place in what seemed reasonable timeframes, go nuts and quasi-Neolithic (some people will be happy with this but I am not, sincerely).
The real problem is that while Y-DNA branching is scattered along time (branches begin low in the tree and continue until very high), mtDNA branching is concentrated at the M-N-R starlike explosions' phase and some derived, yet old, equivalents like the H and U ones. Y-DNA has some of that (D/F-K explosions) but otherwise shows very "recent" lineages like R, E1b1, Q, N and O, massively distributed through the globe.
This is getting too long but I can't resist this few:
"not one but three clades, one of them sister of W".
I'm fairly sure you specifically said that clade does not occur in New Guinea.
"I think my example explains drift pretty well".
No it doesn't. But we can use your beans to illustrate the differences between genetic drift, founder effects and bottlenecks. Starting with a box of multi-coloured beans: gentic drift is the equivalent of taking a bean out every so often, founder effect is the equivalent of taking out just a handful and starting again with them, bottleneck is the equivalent of destroying most of the beans. Human migration into new regions probably involves all three phenomena, but certainly founder effect would be the most important.
You obviously believe there was a single, sudden, rapid expansion of a group of modern humans containing vitually all modern haplogroups. Yet we know from other migrations, such as the Polynesian one, that the three effects mentioned above are extremely important.
"but this 'coupling' pose some other challenges".
Quite. And therefore almost certainly wrong.
I'm fairly sure you specifically said that clade does not occur in New Guinea.
In Australia... Connection with South and West Asia... Reread!
Starting with a box of multi-coloured beans: gentic drift is the equivalent of taking a bean out every so often, founder effect is the equivalent of taking out just a handful and starting again with them, bottleneck is the equivalent of destroying most of the beans.
Depends on your point of view: my bag of beans was always the end result (i.e. I want to know what do I have left inside after I do the respective operations), so no difference between bottleneck and founder effect.
But my example does explain drift:
- You have 100 beans, five colors
- You take out 20 beans randomly (at some number of generations 20% of people has no effective descendace)
- Rinse and repeat.
If all colors are even at the beginning you'd will surely end with all them at the end unless you go extreme taking out every bean until there's only one or two.
But now do with just 10 beans of 2 or three colors: if you repeat the same number of times as before, fixation will happen.
So the original population's number and diversity (founder effect) does influence drift, of course. But it's not so much founder effect (diversity) as founding numbers and population growth/stagnation. If the populaton grows: all grow, you can take no or very few beans out of the bag. But if population is stagnant or receeding, then please let the drift begin.
One of the effects of this process is that in expansive conditions mother clades (por any other old clades) tend to "reabsorb" (drift out) derived ones (minority by default at first), so only lineages that get "out of the bag" can really have founder effects of note... unless something odd happens.
That's why you see R most diverse in South Asia but relatively weak numerically in comparison with other regions where it got much better opportunities. R had some success in South Asia, mark of its still unexplained vigor, but would it be only for that, it would have remained as just another N subclade.
Of all those 6 R subclades that appear to expand early on, four seem to do it in South Asia (R2'JT, R6'7, R30 and R31), while only two do outside (R9'21'22'F and P).
You already see some clades migrating eastward at this first moment of R secondary expansion - and surely others are doing the same, lurking with these but not yet expanding. In the second moment though, nearly all expanding lineages branch out not anymore in South Asia but outside: R0, U (still in South Asia?) and JT (secondary branch) do sequentially in West Asia, while in East Asia it's B and R9'F (secondary branch) who do the same.
So, while R probably found some room for its early expansion in South Asia, this was too constrained for its many energies, and therefore several lineages head in West and East directions, surely following the trail of older M and N migrations.
The major South Asian R subclades are:
- R2 (sister of JT): mainly in the NW, towards Pakistan
- R5: mainly in SW and Bengal
- R6: mainly in the SE
- R7: mainly in the NE (lower Ganges)
- R8: mainly in the SE
- U: partly overlapping with R2 (NW) in the form of U2 and U7. Also in SW and middle Ganges (as well as outside of SA towards the NW).
- R30: NW (Rajastan)
- R31: Sri Lanka especially (and also Australia).
So by concentration (and known Paleolithic routes), the first instinct is to look to the NW for its origins but, as it seems obvious that it had such a big impact in East Asia and Melanesia, with the Western expansion appearing as later in time, I'd say that the real core probably was near Bengal or Bihar.
You obviously believe there was a single, sudden, rapid expansion of a group of modern humans containing vitually all modern haplogroups. Yet we know from other migrations, such as the Polynesian one, that the three effects mentioned above are extremely important.I can't think of any other reasonable explanation. As you know mtDNA is in general much better conserved and there are only 2 such clades in Eurasia. If they would have arrived by spearate routes, having distinct expansion times and geographies we would see that difference very clear and we just don't.
Polynesian expansion can't compare because it has multiple origins, genetically speaking. When humans moved from Africa into Asia, there was no such other genetic pools to tap from, as Polynesians did. They only had (and have transmitted) what they brought with them.
In a sense is much more directly comparable to the racist European colonial expansions of recent centuries, like those of North America, Argentina, Australia or New Zealand, where the native input is negligible. The main difference is that in the recent cases there were many different waves of colonists from diverse sources, as communications were much more advanced, but in general you can see an explosion of European Y and mt lineages, some much more marked, much like you can see an explosion of African Y and mt lineages, especifically those that actively participated in the "paleocolonial" expeditions. It seems these were basically three Y-DNA and two mtDNA lineages. That's it.
Correction (Indian R main locations):
- R2: Rajastan and Karnataka [NW]. Related JT: mostly West Eurasia.
- R5: Bengal, Karnataka and Kerala
- R7: Bihar, Chattisgarh, nearby areas [low Ganges]. Related R6: Tamils and Kashmir.
- R8: Rajastan and South.
- R30: Punjab, Gujarat and Pakistan [NW].
- R31: Rajastan and Sri Lanka. (Also Australia).
- U subclades:
--> U2i: Uttar Pradesh, Kashmir, Sindh, Karnataka and Sri Lanka. (Also SE Central Asia).
--> U7: both Punjabs and Gujarat [NW].
You could well argue for a double core in the lower and upper Ganges for this clade, yet, the expansion to East Asia and Melanesia seems equally. So unsure. I feel that the same happens with M: while the South Asian early expansion seems concentrated in the NW of the subcontinent, its clear push eastward into SE Asia and Sahul, makes it hard to place its origin in the Upper Ganges or Indus.
"In Australia... Connection with South and West Asia..."
I'll check it out soon but still no mention of New Guinea, which was the point of my comment.
"my bag of beans was always the end result".
What's the point of starting with the end result if you're considering drift etc. Surely how you arrive at the end result is what's important.
"I'd say that the real core probably was near Bengal or Bihar".
Or further east.
"I can't think of any other reasonable explanation".
Because you're not prepared to consider any other.
"If they would have arrived by spearate routes, having distinct expansion times and geographies we would see that difference very clear and we just don't".
Yes we do. Most haplogroups have a reasonably restricted distribution. And it's an unjustified assumption that all haplogroups expanded together.
Found it:
"In the mtDNA side of things, instead, Australia does share some derived lineages with South Asia and West Eurasia (but does not seem to share them with East Asia, curiously enough). These are:
· M14 (Australia and Arabia)
· N2 (N2a in Australia, W in South Asia and West Eurasia)
· R31 (in South Asia and Australia)
Notice that all mtDNA connections are with Australia, not New Guinea".
What's the point of starting with the end result if you're considering drift etc. Surely how you arrive at the end result is what's important.
You start with an original population (bag) and take out randomly what is discarded by drift, founder effect or whatever other "bottleneck-like" phenomenon. What is left is the end result and, the way I though of it, it remained always in the bag.
Just a matter of viewpoints anyhow. You can do it inversely or use two bags or whatever.
Or further east.
Hardly. Further east would not explain the highest diversity being in South Asia.
In fact, considering the earchaeological record, the Son-Lower Ganges area is a weaker candidate than the West and maybe even the South of the subcontinent.
I picked the Bihar-Bengal area because of proximity to SE Asia and Sahul but a deeper analysis could well suggest a Western South Asian and/or a Southern South Asian core and probably a strictly coastal route as well.
I'm chewing on it but, in any case, the highest top-level diversity (by far) for both M and R is west of the Ganges Delta.
"I can't think of any other reasonable explanation".
Because you're not prepared to consider any other.
No! Because diversity says so. I am not the one here with the lightbulb-over-the-head syndrome, believe me. It is you who have a prejudiced position and won't concede to the overwhelming factual evidence against it.
Yes we do. Most haplogroups have a reasonably restricted distribution. And it's an unjustified assumption that all haplogroups expanded together.
Only derived haplogroups have (more or less) regionalized distributions and that is just normal and expected. Macrohaplogroups are too widespread and diverse to bend to that pseudo-reasoning of you. And They all share the same basic distribution axis through Southern Asia (i.e. "The Indies").
Why do you talk of "most haplogroups" without any sort of detail or precission? Why don't you discuss mtDNA M or mtDNA R specifically? Or whatever other lineage that you consider relevant? Because it is in the specifics where the truth lays.
"In the mtDNA side of things, instead, Australia does share some derived lineages with South Asia and West Eurasia (but does not seem to share them with East Asia, curiously enough). These are:
· M14 (Australia and Arabia)
· N2 (N2a in Australia, W in South Asia and West Eurasia)
· R31 (in South Asia and Australia)
Notice that all mtDNA connections are with Australia, not New Guinea".
Yes. I am perfectly aware of that. But it means that somehow people was migrating between Australia and South/West Asia "directly" (well not really possible, right?) at realtively late times. Probably this mistery will be solved when SE Asian M*, N* and R* are analyzed with more detail because it can't be that they crossed the Indian Ocean with Paleolithic canoes nor that they left no traces in between (SE Asia).
"You start with an original population (bag) and take out randomly what is discarded by drift, founder effect or whatever other 'bottleneck-like' phenomenon".
That would be a good idea, but it's not what's relevant here. As the population expands the important beans are the ones you're taking out, they're the ones that move, not the ones that remain behind.
Your inconsistency is amusing. One minute you're arguing that Paleolithic populations were small, a position I agree with. Then you claim that virtually all modern haplogroups emerged from Africa within just one original, presumably very small, population. You're then forced to make up some story about how this small Paleolithic population was too stupid to simply move back inland with an advancing coastline as sea level rose, leaving no trace of the route they took from Africa to India. Surely that's unlikely.
"Further east would not explain the highest diversity being in South Asia".
Diversity does not indicate point of origin, as you are very well aware, viz. your recent comment regarding the haplogroup diversity in Australia not actually indicating human origin there. Because today we find a variety of haplogroups in mamy regions you're assuming they've all arrived together, whereas we know that the haplogroup variety in Europe, for example, is a result of a series of immigrations.
"Four additional subclades ... were found to have the
derived state at P14 and the ancestral state at mutations that
define the G-T lineages".
This looks suspiciously as though haplogroup F (in the strict sense) separated from the remainder of the G-T lineage at an early stage of their (separate?) expansions. The most interesting consequence of this would be that Y-haps in the G-T lineages need never have been into India at all, not until K appeared anyway. And Y-hap IJ need never have entered India with Y-hap K (and all the haplogroups derived from it). If we take this possibility seriously and look at the modern distribution of these F(xGHIJK) haplogroups we find that they are distributed in order from G in the Caucasus, J in the Zagros, I north of the Zagros and H probably in Baluchistan. E is found in the Arabian Peninsular and in Africa, but I'll leave you to consider Y-haps D and C. It would make sense that the modern human population became established through this region before entering India.
"Why don't you discuss mtDNA M or mtDNA R specifically?"
The derived haplogroup mtDNA R is especially interesting. P is found virtually only east of Wallace's Line. Any examples found to the west probably represent relatively recent movement west. And F and B are found north of Wallacea, in East Asia. It's only once we look at mt-haps HV, KU and JT that we are forced to consider a more westerly origin.
"But it means that somehow people was migrating between Australia and South/West Asia 'directly'".
True. But not necessarily in the Paleolithic.
That would be a good idea, but it's not what's relevant here. As the population expands the important beans are the ones you're taking out, they're the ones that move, not the ones that remain behind.
Depends how you percieve it, LOL. Have you ever used one of those old lifters without security doors? Are you going up or the world going down? Are you taking beans off the original or the destiny population?
One minute you're arguing that Paleolithic populations were small, a position I agree with. Then you claim that virtually all modern haplogroups emerged from Africa within just one original, presumably very small, population.
Where is the contradiction? The highest diveristy within L3 is in Africa and there's were all upstream relatives are as well. M and N are L3. It's extremely simple.
You're then forced to make up some story about how this small Paleolithic population was too stupid to simply move back inland with an advancing coastline as sea level rose, leaving no trace of the route they took from Africa to India.
Don't be an ass and do not put words I never said or ideas I never even thought in my mouth.
Sea level rising, at least the major one, happened only many tens of thousands of years after the OOA and all that. What we'd like to find under the sea are not genes (as you say well: they moved inland) but for stones and bones: for archaeological evidence.
Diversity does not indicate point of origin...Or does it? If there is migration from multiple origins, then not, but if all, as in the case of a single haplogroup must come from a single origin then, with all likehood yes.
To question the diversity=origin logic you need at least a good reasoning of why.
...as you are very well aware, viz. your recent comment regarding the haplogroup diversity in Australia not actually indicating human origin there.
Not all humans, only the N subclade. Why do I do that?
1. Because we lack of sufficient data from M* and N* meaning in SE Asia.
2. Because we can't be sure that those mostly new N clades described in Australia are not (as have happened elesewhere) a single subclade in fact.
3. Because if it's hard to imagine crossing into Australia with mere canoes, it's even harder to imagine them mass-migrating in the opposite direction.
I considered N as being Australian for analytical purposes and then discarded it for the final result because I was anything but convinced.
"Four additional subclades ... were found to have the
derived state at P14 and the ancestral state at mutations that
define the G-T lineages".
This looks suspiciously as though haplogroup F (in the strict sense) separated from the remainder of the G-T lineage at an early stage of their (separate?) expansions.
Not sure where you took that quote from. But P14 is definitory of F (along with a long list of other markers), so it's shared by all F subclades, numbered or lettered.
Can you realize that nomenclature is nothing but a convention? That F1 is "brother" of G or IJK? That all them are F, the same that all branches from a tree are the tree?
The derived haplogroup mtDNA R is especially interesting. P is found virtually only east of Wallace's Line. Any examples found to the west probably represent relatively recent movement west.
Uh?
P is the only R subclade (the only one!) east of your fetish Wallace line.
I'd ask you why are you singleing it out as of "special" relevance, being actually "the odd man out".
It's only once we look at mt-haps HV, KU and JT that we are forced to consider a more westerly origin.
And all the Indian R clades, which are many.
You are blind to the data that does not have a big capital-letter nomenclature attached, ignoring that such nomentaclature is obsolete and fails to describe properly the South Asian diversity. You insist in being blind and you keep going back over stuff we have already discussed.
You keep putting the cart before the horses and it's therefore going nowhere, obviously.
"But it means that somehow people was migrating between Australia and South/West Asia 'directly'".
True. But not necessarily in the Paleolithic.
Yes because the splitting nodes, the point where one "sister" headed to Australia and the other westward, are very old, most of them surely older than the colonizaton of Europe and that means Middle Paleolithic by the usual age terminology.
"Are you taking beans off the original or the destiny population?"
When talking about founder effect you're obviously considering the ones you've taken off the original. Only some members of this original population are likely to migrate.
"M and N are L3. It's extremely simple".
Exactly. In fact this is very compelling evidence for a major bottleneck, founder effect, call it what you will, in the original OOA movement. Of the many mt-haps already present in Africa at the time just two branches, out of at least seven within just the mt-hap L3 clade, made it even as far as SW Asia (taking into consideration the possibility that the few mtL lineages found in SW Asia may date to that period).
With the Y-chromosome the bottleneck is less obvious, but the four main haplogroups present outside Africa derive from just one branch of haplogroup B. Therefore it's quite possible the bottleneck allowed just one Y-hap to escape from Africa. The four then diverged outside that continent.
"Not sure where you took that quote from".
From one of your posts of 21st April: (Kivisild et al. 2003; Karafet et al. 2005; Sengupta et al. 2006; Zerjal et al. 2007)". Satisfied?
"But P14 is definitory of F (along with a long list of other markers), so it's shared by all F subclades".
P14 is shared by all the F clades but the point the authors were making is that the F clade, in the strict sense, does not share the downstream mutations of the other clades. So the F clade as a whole actually breaks into two. That means that F1 is not a 'brother' of G or IJK, but more a cousin. I suspect that just the F branch actually made it into India. Haplogroup K moved in later.
"P is the only R subclade (the only one!) east of your fetish Wallace line".
OK. So how did it get there? You cannot form a valid scientific theory without considering ALL the evidence, even though it is 'the odd man out'".
"You keep putting the cart before the horses and it's therefore going nowhere, obviously".
But for some strange reason you're not prepared to consider the obvious possibility that ancient haplogroup distribution is a product of much the same processes as we know have influenced more recent haplogroup distribution.
"such nomentaclature is obsolete and fails to describe properly the South Asian diversity".
And that S Asian diversity is a compelling evidence in favour of the idea that humans spent a long time there. There was no 'rapid' migration through S Asia.
"most of them surely older than the colonizaton of Europe and that means Middle Paleolithic".
I've been trying to get through to you for ages that a lot had happened long before modern humans were able to get anywhere near Europe.
When talking about founder effect you're obviously considering the ones you've taken off the original. Only some members of this original population are likely to migrate.
Not so obviously. I was considering the bag's contents as the migrating population all the time.
But, as I said, a matter of perspective. Not worth arguing about it.
In fact this is very compelling evidence for a major bottleneck, founder effect, call it what you will, in the original OOA movement.
Yes and not so much: L3 has 7 top-level lineages, of which two (29%) migrated to Asia. That's a lot for the ofspring of an African lineage.
When you look at the Y side of things is even more confusing: 3/4 (or 1.5/2 if you wish) of all CF'DE lineages migrated to Asia. But well, I'll try t focus in mtDNA.
Still for a time (4 or 5 SNPs) after the African L3 expansion, M and N did not branch out, what means that the population was small and mutations out of the main survivor lineages were "reabsorbed" (drifted out) easily.
This still a shorter span than what you find in other haplogroups' stems. L3f (an important African lineage, directly derived from L3, as M and N) is aparently contemporary in its expansion with N. Under the N and M nodes there are many clades that have longer stems - examples: M1 (10), M2 (7), A and X (8).
It certaily suggests that the M/N population (assuming they were just one) had a period when they merely survived and did not expand. This is accounted for in the coastal migration model by the presumed South Arabian phase, when they were surely just small bands of coastal dwellers in Yemen and Oman. Another possibility would be a true massive bottleneck caused by the Toba catastrophe but it's not a necessary explanation.
Of the many mt-haps already present in Africa at the time just two branches, out of at least seven within just the mt-hap L3 clade, made it even as far as SW Asia...
It is only logical: they lived in Africa and crossing into Asia (that was mostly deserts at the other side of the sea from their viewpoint) was a challenge, not an easy thing to do. So only a few would do that in any case. But those who did, certainly made a good long-term investment in terms of reproductive success.
With the Y-chromosome the bottleneck is less obvious, but the four main haplogroups present outside Africa derive from just one branch of haplogroup B. Therefore it's quite possible the bottleneck allowed just one Y-hap to escape from Africa. The four then diverged outside that continent.
Actually not: haplogroup B is an African-only branch of BT. The other branch is Y(xA,B), CF in the current ISOGG tree, or, as I prefer to call it, CF'DE. This lineage is distinct from B (M60, M181, P85, P90) and defined by M168, M294, P9.1 from the Y(xA) or BT node.
The surprising thing in this lineage is that most derived subclades are Asian, not African. So no apparent OOA bottleneck but a problem to speculate about in fact.
From one of your posts of 21st April: (Kivisild et al. 2003; Karafet et al. 2005; Sengupta et al. 2006; Zerjal et al. 2007)". Satisfied?.
Not really because the especific source is not clear (four different papers). The claim you make (that there is a distinct mutation leading to G, H and IJK within F) is too daring not to be investigated properly. Moreso when it is contradictory with what is found at ISOGG.
I will abide by ISOGG, as it is a most up-to-date reference. Following it: F1, F2, F3, F4, G, H and IJK are sister clades within F.
P14 is shared by all the F clades but the point the authors were making is that the F clade, in the strict sense, does not share the downstream mutations of the other clades. So the F clade as a whole actually breaks into two. That means that F1 is not a 'brother' of G or IJK, but more a cousin. I suspect that just the F branch actually made it into India. Haplogroup K moved in later.
By ISOGG, F1 does share all the F mutations, including P14, like all other F subclades and F* as well.
"P is the only R subclade (the only one!) east of your fetish Wallace line".
OK. So how did it get there?.
Via SE Asia and a boat. What's the problem? Other clades did before her.
But for some strange reason you're not prepared to consider the obvious possibility that ancient haplogroup distribution is a product of much the same processes as we know have influenced more recent haplogroup distribution.
Blanket statement that I do not really understand, sorry.
And that S Asian diversity is a compelling evidence in favour of the idea that humans spent a long time there. There was no 'rapid' migration through S Asia.
Well, SE/E Asian and even Sahulian diversity are evidence of rapid migration along the IP Y. If you look at the mtDNA tree or at my more recent series of maps (easier to see maybe), there are subclades popping like mushrooms in all that area soon after the M explosion. That means rapid migration through South and SE Asia.
Admittedly they seem to have taken a little longer to arrive to Sahul - but just a bit longer, nothing more.
I've been trying to get through to you for ages that a lot had happened long before modern humans were able to get anywhere near Europe.
Well, you don't need to insist on that: I am perfectly aware, I have been always perfectly aware of that fact. The colonization of Europe anyhow is a well dated episode and therefore is a relevant reference point.
"I was considering the bag's contents as the migrating population all the time".
But the migrating population is extremely unlikely to contain exactly the same ratio of haplogrouops as the original. In fact it can contain just a very small subset of it.
"It certaily suggests that the M/N population (assuming they were just one) had a period when they merely survived and did not expand. This is accounted for in the coastal migration model by the presumed South Arabian phase".
There is absolutely no evidence for a 'South Arabian phase'. The population could have lived anywhere on the Arabian Peninsular between Africa and the Zagros Mountains. They are just as likely to have emerged from Africa via the Eastern Mediterranean coast as the South Arabian coast.
'Actually not: haplogroup B is an African-only branch of BT".
OK then. Just one branch of BT escaped from Africa. Is that better? Anyway, an 'apparent OOA bottleneck'.
"the especific source is not clear".
Karafet et al (2008):
http://www.familytreedna.com/pdf/Karafet-et-all-GR508.pdf
"Admittedly they seem to have taken a little longer to arrive to Sahul - but just a bit longer".
But after N, and that's probably significant.
But the migrating population is extremely unlikely to contain exactly the same ratio of haplogrouops as the original. In fact it can contain just a very small subset of it.
True.
I was anyhow focusing only on an arriving population and the possible results depending on what number of beans and variety of them arrived (regardless of the ancestral population).
But it doesn't matter as long as you see it clear.
There is absolutely no evidence for a 'South Arabian phase'.
True. It's just a logical deduction, as it's the shortest route and presumably all the rest was under Neanderthal control.
The population could have lived anywhere on the Arabian Peninsular between Africa and the Zagros Mountains. They are just as likely to have emerged from Africa via the Eastern Mediterranean coast as the South Arabian coast.
Probably not because of two reasons:
1. The Fertile Crescent, and even Egypt before that, should have supported an autonomous population expansion of which there are no traces.
2. If the migration happened as late as the mtDNA seems to suggest (c.60 kya) then the Fertile Crescent was a Neanderthal country already.
But admittedly it is an obscure event. You oppose the simplest explanation because you refuse to admit that people could use boats before arriving to your darling Wallacea. But they could and surely they used them.
In fact I see absolutely no reason for people to have boats in Wallacea c.50 kya and not in Djibouti c.60 or 70 kya.
In any case, arguing on this issue is quite trivial because we have no evidence of anything between the L3 and the M nodes. If they were abducted by ETs at one spot and dropped at the other we could never find out. At least not with the data we have at the moment.
OK then. Just one branch of BT escaped from Africa. Is that better? Anyway, an 'apparent OOA bottleneck'.
2/3 of one out of two branches. If that's a bottleneck, "let God come down and witness it", as they say over here.
BT is trivial here: CF (CF'DE) is what matters here. B branched out at a previous stage. The "bottleneck" (or rather extreme drift) is in E, not C, D and F. As they also say over here: "three is a crowd".
Personally, I'd prefer that Eurasian Y-DNA was simpler. One, maybe two founding fathers would be more than enough considering what we see at the mtDNA level. But reality is stubborn: they were at least three.
Karafet et al (2008)It reads:
"Four additional subclades: F-P91/P104 (F1), F-M427/M428 (F2), F-P96 (F3), and F-P254 (F4) were found to have the derived state at P14 and the ancestral state at mutations that define the G-T lineages".
True.
And you commented: This looks suspiciously as though haplogroup F (in the strict sense) separated from the remainder of the G-T lineage at an early stage of their (separate?) expansions.
Now please read the original quote again. I let myself be driven by your interpretation but it only says that F1, F2, F3 and F4 are as much F (defined by P14) as G, H, IJK, S and T. Not that they are different. Not at all.
"Admittedly they seem to have taken a little longer to arrive to Sahul - but just a bit longer".
But after N, and that's probably significant.
It might be significant. I really don't know.
Just read it geographically and temporarily:
1. South Asia - M event
2. SE Asia - N event
3. Sahul (no particular event associated)
It's just the logical continuation of a process. But whatever arrived to Australia and Papua, basically became stuck in there: there was nowhere else to go beyond and behind was too remote and too densely populated already. So S or M29'Q (or P later on) never had the chance to become the new superclade.
"BT is trivial here".
But it must have existed. If a branch of it was the only haplogroup that emerged from Africa we would indeed have a bottleneck.
"Not that they are different. Not at all".
When the classifiers changed K1 to M2 and K7 to M3 they did it for a reason. K is separate, it's no longer a catchall haplogroup. It shares just some of the mutations that gave rise to its cousins L (in India), T (in SW Asia), S (in New Guinea), M (in Melanesia) and NOP (somewhere). Same with F and haplogroups G (Caucasus), H (Baluchistan) and IJ (somewhere).
I believe we can discern four separate movements across Wallacea. The Austronesian expansion is the most recent but it cannot be considered a separate phenomenon to the other three. Boating technology in the region has been a continuously evolving process.
We seem to agree that the first group of humans across Wallacea belonged to Y-hap C and mtDNA N. In Australia they became Y-hap C4 and mtDNA S. I suspect they crossed between the warm periods you propose of 73K and 59-50 K.
Various lines of evidence (not just DNA) suggest that New Guinea was wasn't occupied till after Australia was, possibly following the rise and subsequent fall of sea level after 50K. At this stage the original Wallacea haplogroups had become extinct on some islands and so outnumbered by the original inhabitants of Sunda. Y-hap S and then M derived from it and mtDNA Q, probably derived from S then arrived in New Guinea and a little way beyond. So the second movement across Wallacea is by Y-hap Fs line along with mtDNA Ms line, not Y-hap C or mtDNA N. Suggests a separate origin.
A third movement into Wallacea occurred when Y-hap K came ripping through and picked up mtDNA P and took her to both New Guinea and Australia. Y-hap C6 may have become involved with either Y-hap S or Y-hap K. It's irrelevant which.
The fourth movement across Wallacea is much later. Y-hap C2 took R's descendant B into the wider Pacific about 5K.
"BT is trivial here".
But it must have existed. If a branch of it was the only haplogroup that emerged from Africa we would indeed have a bottleneck.
It did exist, sure. In fact it still exists in derived form and is the most common Y-DNA haplogroup on Earth actually.
But at that node, the normal explanation is that it branched in two: one lineage went towards Middle Africa and the other towards the Horn of Africa. We can talk of small population overall maybe, of strong drift, but splitting in two is not "a bottleneck" as I see it. And at that stage most people think that all happened within Africa, so prior to the migration to Asia anyhow.
What is considered re. the OOA is CF (or better IMO: CF'DE). Unlike mtDNA L3, CF'DE sent more lineages to Asia than to Africa. I and surely others have some problem understanding this but it's a stubborn fact.
When the classifiers changed K1 to M2 and K7 to M3 they did it for a reason. K is separate, it's no longer a catchall haplogroup. It shares just some of the mutations that gave rise to its cousins L (in India), T (in SW Asia), S (in New Guinea), M (in Melanesia) and NOP (somewhere). Same with F and haplogroups G (Caucasus), H (Baluchistan) and IJ (somewhere).
That's a total nonsense: ISOGG: haplogroup K 2009.
What happened was that those K lineages were found to share many founder SNPs whith M (now M1) and the tree was reorganized accordingly. But M still hangs from K, just that M1 (former M), M2 (former K1) and M3 (former K7) now are known to form a single subset of K and are not different lineages, as was believed before.
This does not affect the rest of K. And certainly L and the rest are not "cousins" of K: they are its sons.
Get that straight please because if you don't understand the basics, you can hardly undrstand the big picture.
I believe we can discern four separate movements across Wallacea. The Austronesian expansion is the most recent but it cannot be considered a separate phenomenon to the other three. Boating technology in the region has been a continuously evolving process.
For me what happened 40-50 kya and what happened just 3 kya ago are very different things. Not just technologically but also conceptually. The other flows into Sahul were rather continuous (even if they had gaps and pulses) through some 10 thousand years or maybe less. Then there is maybe 40 milennia of Sahulian isolation and then come the Austronesians, who are much closer in time and technlogy to the Chinese traders and European colonists.
Taking all that as continuum, you could well include Hanno, Erik the Red, Zheng He and the Caribbean sailors for what I can tell. If Colombus' travels were yesterday, the Autronesian ones were six days ago but the colonization of Sahul in the MP happened three months ago instead. Austronesians and Colombus are obviously more related among them than any of them with the Sahulian aborigines. They are probably by events that happened inland (steel tech for example) but they are still more connected anyhow, not just in the timeframe.
Various lines of evidence (not just DNA) suggest that New Guinea was wasn't occupied till after Australia was...
Well, mtDNA does not suggest that but rather simulatenous colonization.
I'd say that if you want to concieve just two main pulses, the first one was probably (Y)C with (mt)N in Australia but (mt)M in Melanesia, and the second coul be (Y)K with (mt)N (again) and R into Australia but (mt)R only into New Guinea. Not too sure about Y-DNA anyhow because they could also have arrived C and K together and drifted separately in the different areas, gradually eliminating previous Y-DNA (F(xK) remnants?).
Remember that in the Y-DNA tree I worked on earlier, I got C apparently more recent than K, and even more recent than M. And all Australian lineages expanding later. So maybe (Y)M arrived with (mt)M to Melanesia - would make sense.
At the moment anyhow I can only consider mtDNA and Y-DNA separately. I find extremely difficult to couple them properly and have no definitive conclusions, not even preliminary ones about such possible pairings.
"This does not affect the rest of K. And certainly L and the rest are not 'cousins' of K: they are its sons".
Obviousl;y you didn't read the link you provided:
'At present this group [Y-DNA haplogroup K] contains two distinct classes of subgroups: (1) major groups L to T (refer to the main tree at Y-DNA Haplogroup Tree) and (2) minor groups K* and K1 to K4 which do not have any of the SNPs defining the major groups'.
As you say, "Get that straight please because if you don't understand the basics, you can hardly undrstand the big picture".
Yes: they don't have the same SNPs: they have their own defining SNPs. The verbal excercise of someone about "major" and "minor" haplogroups is confusing you. Each different haplogroup has not any of the SNPs that define any other haplogroup within K.
Just check the SNPs, please.
"have their own defining SNPs".
Which is what gives rise to 'two distinct classes of subgroups', separating them from the other related haplogroups. You don't understand that yet, do you?
"Each different haplogroup has not any of the SNPs that define any other haplogroup within K".
Quite. But at the base they divide into two, as does Y-chromosome haplogroup F. You're confusing basal mutations with later mutations. Presumably so you can still claim they all expanded from the same region.
I'll repeat, because it obviously hasn't sunk in, "At present this group [Y-DNA haplogroup K] contains TWO DISTINCT CLASSES OF SUBGROUPS: (1) major groups L to T ... and (2) minor groups K* and K1 to K4".
"Just check the SNPs, please".
I think that would be a good idea> Have a look:
http://www.isogg.org/tree/ISOGG_YDNATreeTrunk08.html
Qué no, hostias!
"have their own defining SNPs".
Which is what gives rise to 'two distinct classes of subgroups', separating them from the other related haplogroups. You don't understand that yet, do you?
.
It's you who don't understand it. Do not look at the words but at the SNPs. The words are confusing you because they are somewhat vague. But they just mean that each K sublineage has its own distinct markers, what is just obvious, not that the different K(xL,M,NOP,S,T) form any unified anything. Each K lineage has its own distinctive markers: same for K4 and for T.
this is the clear one for KPlease...
contains TWO DISTINCT CLASSES OF SUBGROUPS: (1) major groups L to T ... and (2) minor groups K* and K1 to K4"Major and minor = large and small. No other meaning. Vague language and wishful thinking, Terry.
I realise English is not your first language so you probably don't completely understand it. So, just to explain things a little better here are some relevant quotes from the Karafet paper:
"As for haplogroups F and P (see below), lineages within haplogroup K are not united by a single mutation. Instead, haplogroup K is characterized by the derived state at four sites (M9, P128, P131 and P132) and the ancestral state at the mutations that define the L, M, NO, P, S and T lineages (fig. 1). In addition to paragroup (K*), there are five mutations that define four different K lineages"
OK. So what could possibly be similar between K and F. As I've posted before the authors wrote regarding F:
"Four additional subclades: F-P91/P104 (F1), F-M427/M428 (F2), F-P96 (F3), and F-P254 (F4) were found to have the derived state at P14 and the ancestral state at mutations that define the G-T lineages".
Obviously a completely different situation, so the authors don't know what they're talking about comparing K with F. Especially because under P the only thing they write is:
"Haplogroup P consists of two widely distributed Q and R lineages (fig. 1)".
And that's it. Nothing else. Obviously nothing at all in common with either haplogroup F or haplogroup K. And certainly no mention of major and minor lineages, whatever they are.
Well, I still get like 130 in English-language standard IQ tests. I manage well enough with written English, I think and i have been praised for that in many occasions (though sure: now and then I can't find the word, especially if it has a germanic and not romance root).
You should stop reading that paragraph once and again and go to the SNP tree and check what's up with K.
For me the paragraph is clear enough though the rather unscientific distinction bewteen major and minor haplogroups has confused you, by feeding your wishful thinking processes.
When Tatiana Karafet wrote: "In addition to paragroup (K*), there are five mutations that define four different K lineages" (a sentence that I understand perfectely and I could translate to some other two languages without thinking much), she obviously means that K1, K2, K3 and K4 are known as distinct to paragroup K*, or K(xK1,K2,K3,K4,L,M,NOP,S,T). That is of course what is reflected in the ISOGG tree 2009, which is was massively reviewed one year ago or so to adapt to Karafet's findings majorly.
Sincerely get over it and stop using that sentence as a pretext for a blatant error.
Haplogroup P consists of two widely distributed Q and R lineages (fig. 1)".
And that's it. Nothing else.
Because as far as we know there is nothing else. So?
Man: check the tree and stop making a mountain out of an imaginary grain of sand. It's boring and doesn't say much in your favor.
And certainly no mention of major and minor lineages, whatever they are.
There are no minor ones in this case (as far as we know). Anyhow major and minor is just literary blah-blah. One could perfectly argue that T is a minor clade or that K3 should be thought of as major: the tongue confuses the facts. This may be very useful when you're onto propaganda or publicity but can be a total mess when you are trying to describe scientific facts.
As someone wrote: the preciession achieved in names in scientific language is totally destroyed when it comes to verbs. What does a gas suffering a pressure of X atmospheres, for instance?
Literature, language itself, is not concieved primarily to serve science but human communication, which is largely emotional, not just factual. "Major" and "minor" are just emotional, quantitative if you wish, pretty much unscientific categories. The ability of any person (Karafet or whoever) to write a paragraph that coherently transmits the information intended eaully to all minds is relative. Artists realize that and accept that once finished the artwork becomes autonomous by that kind of mental "magic".
But art is not factual information. And literature is art.
In other words: you're hanging from the proverbial "burning nail".
But art is not factual information.
Waddinton's thoughts on the matter-
"Karl Popper pointed out that given the statement 'If p then q' and you do p and q does not follow, then you have disproved the statement; and he argued that the real method of science is not to try to verify statements but to disprove them. A surprising number of scientists, including very successful ones, have expressed agreement with him. But Poppers's argument only holds in the abstract world of pure logic, in which the statement 'If p then q' implies that q will always follow p, whatever all the other circumstances are, so that we are not allowed to bring in some other disturbing factor. ... But the world which science tries to to analyse is not the pure world of logic; it is the rough and untidy world of actual happenings. Suppose we have a hypothesis like 'If a match is put to twigs (p) a fire starts (q)'. Quite often, when one does this (p) the fire actually does start (q). But again , sometimes the twigs are wet, or something; and this does not completely disprove the suggestion that matches have something to do with starting fires. In the real world of science we can never have statements which are 100% true in all circumstances. The mistake made...-- (by) those who asked for verification and those who would settle for falsification-- is that they demand 100% certainty; and that is something we can never have about the real world. All science can do is to show that some things are very likely, others unlikely. Its picture of the world is more like a portrait drawn by a painter than like a precise theorem in logic" (CH Waddington 1977)
OK. Let's change the haplogroups mentioned in the quotes.
"haplogroup C is characterized by the derived state at four sites (M216, P184, P255 and P260) and the ancestral state at [many of] the mutations that define the F, G, H, and IK lineages. In addition to paragroup (C*), there are eleven mutations that define six different C lineages"
"Four additional subclades: F, G, H and IK were found to have the derived state at P14 and the ancestral state at [many] mutations that define the C lineages".
"Haplogroup CF consists of two widely distributed C and F lineages".
I forgot.
"Because as far as we know there is nothing else. So?"
So what's the significance of comparing the split to the situation for Y-haps F and K? Just to fill up the paper?
First example is meaningless because neither C,F is "C", nor H is parallel to F, but a descendant, nor C* is a haplogroup (but a paragroup). C,F (old CF) is comparable in its structure to P, IJK or NOP: one trunk and two, only two, branches (which branch out in turn, of course).
So what's the significance of comparing the split to the situation for Y-haps F and K? Just to fill up the paper? .
She's NOT comparing F with K at all. She is describing them in different chapters using different words.
This argument is irelevant anyway. I'd like to point out that if each K is just another single branch equal to T, L, N-R, S and M then, like these related haplogroups, all the Ks simply sprang up along the route between India and Australia/Melanesia along with the other above haplogroups. And it frees up T from the need to have ever been anywhere near India at all. In my diagram the T of line K simply moves outside the dashed line representing India.
I'd like to point out that if each K is just another single branch equal to T, L, N-R, S and M then, like these related haplogroups, all the Ks simply sprang up along the route between India and Australia/Melanesia along with the other above haplogroups.
Yes, I agree with this.
And it frees up T from the need to have ever been anywhere near India at all. In my diagram the T of line K simply moves outside the dashed line representing India.
T as such doesn't seem to have any meanigful presence in South Asia. But considering where the other K-derived lineages are found, it's very much likely that pre-T (stem stage) was at some stage there.
Post a Comment