PDA

View Full Version : "Craniometrics Reveal “Two Layers” of Prehistoric Human Dispersal in Eastern Eurasia"



Eterne
02-24-2019, 01:21 PM
https://www.nature.com/articles/s41598-018-35426-z

This new paper got some press this week. From an adna perspective, huge dataset of craniometric measurements of Siberian, East Asian and Sahul populations.... but the scenario they present of separate dispersal of proto-East Asians into Siberia from West Eurasia, then isolated for 41kya there(!) is flatly contradicted by ancient dna showing much more recent arrival from the south of fairly fully formed East Asian ancestry (https://www.biorxiv.org/content/10.1101/448829v1 - "The first inhabitants, a previously unknown population of "Ancient North Siberians" (ANS), represented by Yana RHS, diverged ~38 kya from Western Eurasians, soon after the latter split from East Asians. Between 20 and 11 kya, the ANS population was largely replaced by peoples with ancestry from East Asia, giving rise to ancestral Native Americans and "Ancient Paleosiberians" (AP), represented by a 9.8 kya skeleton from Kolyma River. Between 11 and 4 kya, AP were in turn largely replaced by another group of peoples with ancestry from East Asia, the "Neosiberians" from which many contemporary Siberians derive.")

All that Siberian cranial dataset is from that post 20kya and even largely post 11kya expansion... The perils of trying to estimate ancient continuity given massive post-glacial replacement, followed by ongoing late Holocene expansion...

They suggest alternatively: "In terms of the deeper origins of the apparently homogenous NEA population, we may consider the more ancient homelands and migratory routes, prior to the entrance into the Yellow and Yangtze River areas by 9 kya but potentially much earlier. In one possible scenario, ancient people perhaps of the “first layer” with Australo-Papuan features moved into Siberia and subsequently adapted to the extremely cold climate during the Last Glacial Maximum (LGM) of 24– 16 kya.". But that seems a bit strange again in the light of the results from the Yana paper (even when East Asian related ancestry emerges in Siberia at Kolyma at 8000 BCE and then outside Siberia at Devil's Gate at 5000 BCE, there's not much indication that the early Kolyma like groups back-migrated south again, as there's a lack of their Ancient Siberian related ancestry).

Another thing I wanted to comment on is the absence of pre-neolithic material from North China plain, and northern precincts generally. Does that reflect a real absence of existing material (prior to the neolithic HG just weren't living there) or a sampling gap?

TuaMan
02-24-2019, 03:54 PM
Craniometrics definitely seems hit-or-miss when it comes to inferring actual population history when cross-validated against actual genomic data. I guess a topical example would be the genomic analysis that was recently done on this Upper Paleolithic Mongolian skullcap: http://www.ox.ac.uk/news/2019-01-30-ancient-mongolian-skull-earliest-modern-human-yet-found-region

Prior to the sequencing, many suspected this was an archaic hominid of some kind based on the morphology of the calvarium, and I can definitely see why - very prominent brow ridges with a very low skull and almost no forehead. Although to be fair, I think so far they've only been able to extract mtDNA from the skull, so if they're ever able to get a whole genome sequence out of it, this specimen might still end up having a significant proportion of archaic ancestry that could explain the unusual features it exhibits relative to modern humans.

Kristiina
02-25-2019, 08:46 AM
But that seems a bit strange again in the light of the results from the Yana paper (even when East Asian related ancestry emerges in Siberia at Kolyma at 8000 BCE and then outside Siberia at Devil's Gate at 5000 BCE, there's not much indication that the early Kolyma like groups back-migrated south again, as there's a lack of their Ancient Siberian related ancestry).

Another thing I wanted to comment on is the absence of pre-neolithic material from North China plain, and northern precincts generally. Does that reflect a real absence of existing material (prior to the neolithic HG just weren't living there) or a sampling gap?

Yana is not everything. The first Siberian genome is Ust Ishim 45 kya and it is more related to modern East Asians than West Eurasians but it seems to precede East-West separation. Tianyuan is also interesting. It is related to Goyet in Europe and Ust Ishim. Yana is 15 kya younger than Ust Ishim.

The lack of any ancient autosomal data from China (apart from Tianyuan) is a great pity.

The tree diagram from the Dzudzuana paper is interesting, but we have to keep in mind that it lacks Onge-like reference.

2905329052

Eterne
02-25-2019, 09:59 AM
@Kristiina, you're correct Ust Ishim is really important to our understanding, but to say that Yana isn't everything in Siberia because of him is a bit like saying Kostenki and all post Aurignacian Europeans aren't everything in Europe because of Oase1. UI has no certain contribution to any later population at this point (no real evidence for it, and some evidence against it in the general equal relatedness of later post-Basal populations).

(Though in the supplements to Sikora 2018, Ust Ishim does actually have his highest formal stats with Yana, funnily enough, so if his pop contributed to anyone sampled so far, it's probably Ancient North Siberians and Ancient North Eurasians, as would seem most probable.)

There were surely phylogenetically ENA populations who contributed to both East Asians and also increasingly to later Ancient North Eurasians through time though. Just not much sign so far of what the authors here posit; an early split between their "two layers" leading to East Asians and all other ENA, where one layer migrated through Central Asia and was resident in NE Siberia for upwards of 40000 years. Instead the latest inhabitants of Siberia from the late LGM onwards and mid-Holocene look like arrivals from somewhere within the broader East Asian region, and inference from modern samples is misleading the authors of this cranial study.

The phylogenetic position of Dzudzuana is going to be important as well, like you say.

Kristiina
02-25-2019, 10:32 AM
UI has no certain contribution to any later population at this point (no real evidence for it, and some evidence against it in the general equal relatedness of later post-Basal populations).


It is the same with Yana, Sikora et al could not make Yana to contribute to modern populations. We know from the uniparental data that there is no yDNA P(xQ) or mtDNA U2’3’4’7’8 in Northeast Asia or America.

29055

Kristiina
02-25-2019, 12:36 PM
Just not much sign so far of what the authors here posit; an early split between their "two layers" leading to East Asians and all other ENA, where one layer migrated through Central Asia and was resident in NE Siberia for upwards of 40000 years. Instead the latest inhabitants of Siberia from the late LGM onwards and mid-Holocene look like arrivals from somewhere within the broader East Asian region, and inference from modern samples is misleading the authors of this cranial study.


In addition to this new paper, we have the following:

Uniparental markers: K2a (Ust Ishim) and K2b (Yana, Malta) and mtDNA R (Ust Ishim), Tianyuan (R/В), Yana (R/U2’3’4’7’8), Malta (R/U*) in Siberia and northern East Asia.

Classical genetic markers: Computer simulation of human leukocyte antigen genes supports two
main routes of colonization by human populations in East Asia
29058

https://www.google.fi/url?sa=t&rct=j&q=&esrc=s&source=web&cd=10&ved=2ahUKEwj6w5bS9NbgAhXw2eAKHXTKDrgQFjAJegQIBRAC&url=https%3A%2F%2Farchive-ouverte.unige.ch%2Funige%3A77111%2FATTACHMENT01&usg=AOvVaw2CJilNispKxLzmZvg2DCvV

Male Demography in East Asia: A North–South Contrast in Human Population Expansion Times
"We have investigated male demography in East Asia, applying a Bayesian full-likelihood analysis to data from 988 men representing 27 populations from China, Mongolia, Korea, and Japan typed with 45 binary and 16 STR markers from the Y chromosome. According to our analysis, the northern populations examined all started to expand in number between 34 (18–68) and 22 (12–39) thousand years ago (KYA), before the last glacial maximum at 21–18 KYA, while the southern populations all started to expand between 18 (6–47) and 12 (1–45) KYA, but then grew faster. We suggest that the northern populations expanded earlier because they could exploit the abundant megafauna of the “Mammoth Steppe,” while the southern populations could increase in number only when a warmer and more stable climate led to more plentiful plant resources such as tubers."

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1456369/

Kristiina
02-25-2019, 12:37 PM
deleted

Megalophias
02-25-2019, 08:41 PM
Zhoukoudian is only about 10 km from Tianyuan Cave. Though the date of the Upper Cave occupation is uncertain (this paper says 34-10 ka BP) the most recent study (https://www.ncbi.nlm.nih.gov/pubmed/29778246) I found says minimally 35.1-33.5 ka, which would put these samples not too far from Tianyuan in time as well.

I don't remember what the phylogenetic position of Tianyuan is (or if that was ever really sorted out). Can the Northeast Asian cranial type be a specialization of a more generalized Australasian type?

Eterne
02-25-2019, 09:44 PM
@Kristiin: No, but Yana is a closely related clade to populations which do, across Siberia until the early post-glacial and then mid-Holocene replacement. Ust Ishim does not seem to form a clade with anyone else we can detect (and certainly not any present day East Asian or Siberian population).

Yana and its clade don't contribute to modern populations in Siberia much because of <20kya expansions and possibly extinctions due to climate. Ust Ishim may not because of post-Ust Ishim, pre-Yana expansions of ANS for the same reason.

On the other points, respectfully of your knowledge here, I don't really believe think y-str from 2006 or HLA from 2015 on sets of modern populations are going to find anything much different than autosomal and cranial dna on the same sets of population failed to find, and I guess I've discussed the reasons why they failed to find what actually happened. Routes that go over Mal'ta and Afontova Gora near Baikal as in those maps don't really work for East Asia, because the people who we samples of living there don't form a clade with them. That is, however much they analyse it, the available set of modern data only includes truly massive expansions and replacements from the south, not anyone actually representative of people who lived there 20-11 kya, and not really even of people who lived there pre-11 kya to a great extent.

Kristiina
02-26-2019, 06:26 AM
I am interested in hard data.

Whatever we propose, we have to take into account the K2 node of the yfull tree.

29062

Yana P is slightly younger than the TMRCA of P, while Ust Ishim K2a is slightly older than the TMRCA of K2a on yfull. K2a and K2b formed 45400 years ago, which is more or less the age of Ust Ishim.

Eterne
02-26-2019, 10:07 AM
OK, so what does that imply for you about the dispersal patterns under discussion, and why?

Kristiina
02-26-2019, 10:24 AM
K2 could have arisen in India or in South Central Asia. K2a probably arose in South Central Asia while K2b may have arisen in India. K2a reached Siberia 15 kya earlier than K2b.

As for archaeology, you should read this: "Emergence and Diversity of Modern Human Behavior in Paleolithic Asia".

"It is possible that some of the earlier, simple core- and- flake- tool industries from China were produced by early modern humans who inhabited the region before the introduction of blade technology. Still, the earlier appearance of various forms of modern traits in southern Siberia than in East Asia, as well as the generally accepted Siberian origin of northern Chinese blade technology (Brantingham et al. 2001; Derevianko 2011; Li et al. 2013; Zwyns 2012), are difficult to accommodate without supposing a northern route of modern human dispersal. The early modern human groups following the northern and southern routes eventually met and interacted with each other, probably in East Asia, as some previous researchers have discussed (Bae 2010; Kato 1996; Oda 2003). Archaeologically, the spread of blade technology in East Asia ~38 ka likely reflects migration and / or cultural diffusion from southern Siberia.3"

In any case, we have very little ancient DNA from the relevant area and, therefore, we have a lot to discover and learn.

Eterne
02-26-2019, 10:47 AM
There are probably a lot of scenarios that are compatible with the technological diffusion of those paleolithic industries and the present day y-dna distribution that don't imply any long term residence of any group providing the vast majority of ancestry present-day Siberians in Siberia longer than 20-11 kya or support a migration route for their main ancestral trunk through Siberia (which both are what adna appears to show unambiguously to not be the case).

The only clear indication that we have that any particular y group reached Siberia at any time is in ancient individuals from that region, and this is currently sparsely represented.

A northern and central route of dispersal certainly happened, the question is the extent to which it is ancestral to present day Siberians (apparently not much in East Siberia, and better represented by descendants in the Americas and the Kets, and even then partially, then to a probably lesser extent groups in West Eurasia).

(The absence of a genetic connection between Gravettian and Siberian groups, despite connections being estimated from archaeological connections in their industries, is an example of where we've seen technological diffusion and connection scenarios without much or any shared ancestry, before).

Kristiina
02-26-2019, 03:13 PM
I do not understand why many people are so dead against this idea of a northern route to China.

It also gives a new perspective to the Macro-Altaic language family and could help understand the division between Nostratic north (Turkic, Mongolic, Tungusic, Yukaghir), ambiguous middle (Sino-Tibetan, Japanese, Korean) and polysynthetic east coast (Nivkh, Chukchi-Kamchatkan, Ainu + Native American languages and Papuan languages).

Eterne
03-01-2019, 10:44 AM
Kriistina, missed this a few days ago. I like your introducing the topic of how this interacts with language dispersals. But from my perspective:

1) I am skeptical the high order structure of languages ("Nostratic" etc) at long time depths can be recovered (the comparative method cannot do it as a cross check to this method trying to connect languages based on relatively few items of selected lexicon, all the well-worn criticisms of these deep macrofamily reconstruction hypotheses)
2) even among those who believe in a "Nostratic" family don't place it at a disperal at 45kya(!), usually the epipaleolithic Near East, and did not propose reconstructing languages at the Out of Africa stage
3) languages can become detached from a population genetic component* and older languages have a lot of time for this to happen

This isn't to say that "Nostratic" languages being connected by a population movement won't happen, but I think not by the framework of migrations in this paper (maybe by some later connecting migrations that are only very loosely present in genetic signal). I don't think there is a solid language connection as evidence for or against present day eastern Siberians (and the languages they speak) ancestors having the 45kya long residence in Siberia this paper hypothesizes, or shorter 20-11kya and post 11kya population history there that adna strongly and directly suggests.

Really, the only substantial reason I'm against the population dispersal model they suggest is because the adna doesn't confirm it and seems to disconfirm it.

*Even within IE people on here have argued that the Anatolian IE languages, where there is no steppe component found in places where it is likely IE speakers had migrated to, could have become detached from original population genetic ancestry. Whether this is right or not, so even more so with much older language connections there could be much more opportunity for this to take place.

TuaMan
03-04-2019, 01:11 AM
One thing that I find interesting about Eurasians is the discordance between autosomal heterozygosity and uniparental diversity. East Eurasians are supposed to have less autosomal diversity than West Eurasians with greater numbers of runs of homozygosity, but they seem to blow West Eurasians away in terms of diversity of uniparental lineages, especially in terms of mtDna. M has over 50 primary branches, all of them solidly East Eurasian with the exception of M1.

West Eurasians are pretty much exclusively N derived, but they don't have a monopoly on N lineages by any means. Of the 17 primary branches of N, only 3 look like they could be exclusively West Eurasian (N1'5, N2, and X). Then looking at R, a macro sub-haplogroup of N, we find another 17 primary branches, it looks like only 5 or so could be exclusively West Eurasian (R0, R1, R2'JT, R3, and U).

However, the Y side of things might be a bit more complex. Most subclades of F seem West Eurasian (G, I, J, and probably H), K seems like it's a tweener because although K2 is obviously the father of the majority of East Eurasians today, its sibling LT seems West Eurasian. I personally think C and D represent the oldest layer of East Eurasian Y-DNA before K2 entered the picture.

parasar
03-04-2019, 01:50 AM
I do not understand why many people are so dead against this idea of a northern route to China.

...

IMO that was the only route as far as the AMH are concerned.
An old post: "there is just no evidence from India prior to Orsang. Which means India did not participate in the AMH occupation of SE Asia and Australia as the map depicts. This actually has been known from the very beginning when the OoA coastal theory was proposed - that India is a complete blank for AMH in the proposed period of coastal migration."
https://anthrogenica.com/showthread.php?1483-24-000-year-old-Y-DNA-R-AND-MTDNA-U-FOUND-IN-SIBERIA!!!!&p=17795&viewfull=1#post17795

TuaMan
03-04-2019, 02:32 AM
The whole southern coastal dispersal route honestly seems pretty flimsy. The first sustained "AMH" (which I think is also kind of a non-sensical term) African migration into Southwest Asia starts all the way back to at least MIS 5 from Egypt (the Nubian complex). I think a proto-Eurasian population was already established in Arabia, Mesopotamia, the Persian Gulf, etc., by the end of MIS 5. Then with the onset of hyper-arid MIS 4, they became isolated and experienced a severe population crash (the famous Eurasian bottleneck), and finally emerged out of it with the onset of MIS 3. There's no real hard evidence tying OoA to a crossing of the Bab el-Mandeb and then a dispersal along the Indian Ocean rim.

Eterne
03-04-2019, 07:20 PM
IMO that was the only route as far as the AMH are concerned.

An old post: "there is just no evidence from India prior to Orsang. Which means India did not participate in the AMH occupation of SE Asia and Australia as the map depicts. This actually has been known from the very beginning when the OoA coastal theory was proposed - that India is a complete blank for AMH in the proposed period of coastal migration."

https://anthrogenica.com/showthread.php?1483-24-000-year-old-Y-DNA-R-AND-MTDNA-U-FOUND-IN-SIBERIA!!!!&p=17795&viewfull=1#post17795

To clarify my view on this if of interest, I don't think it's implausible that we could have had some movement through Siberia by proto ENA (pre Tianyuan) very who were largely replaced by ANE, or a separate movement through Central Asia by a separate ENA branch somehow. Early movement through Inner Asian Mountain Corridor or else Siberia plus and subsequent early replacement in Siberia (I guess first Siberians who were replaced later but whose descendants came back far later in time) no problem.

It's the bit where present day ancestors of Siberians are living in Siberia and adapting in Siberia for much that 45kya that seems to be pretty much out; they must have been living somewhere south of Siberia instead, for most of 45kya->20kya and most of the morphological adaptation (craniometric shape etc) happened outside of Siberia.


One thing that I find interesting about Eurasians is the discordance between autosomal heterozygosity and uniparental diversity. East Eurasians are supposed to have less autosomal diversity than West Eurasians with greater numbers of runs of homozygosity, but they seem to blow West Eurasians away in terms of diversity of uniparental lineages, especially in terms of mtDna. M has over 50 primary branches, all of them solidly East Eurasian with the exception of M1.

West Eurasians are pretty much exclusively N derived, but they don't have a monopoly on N lineages by any means. Of the 17 primary branches of N, only 3 look like they could be exclusively West Eurasian (N1'5, N2, and X). Then looking at R, a macro sub-haplogroup of N, we find another 17 primary branches, it looks like only 5 or so could be exclusively West Eurasian (R0, R1, R2'JT, R3, and U).

I don't think that's necessarily a problem to explain; Bayesian skyline plots basically suggest to me that this is explained by larger increases / fewer population crashes in particularly East and South Asian population size between 50-10kya, which seems fairly consistent with those regions being warmer / less arid and having higher carrying capacity during the ice age.

See figures from Karmin 2015 (https://genome.cshlp.org/content/early/2015/03/13/gr.186684.114.full.pdf) - figures (https://imgur.com/a/ZJ59WdM)

It looks like autosomal dna isn't really subject to the same dynamics; subsequent greater increases in population size in South and Eastern Eurasia Asia and restricted population size in Western Eurasia don't do much to reverse the all conquering power of the serial OoA bottleneck.

Of course you can't take these totally literally, because for instance of course the South Asian mtdna pool was not actually necessarily evolving all in South Asia during that time period, and much of it must have subsequently come from West Eurasia, while the Siberian mtdna pool probably entered Siberia relatively late, etc. There are also more tumultous crashes and decreases and increases in reality that will be smoothed out in their curve. And y-dna is worse, more dislocated. But they are suggestive.

TuaMan
03-05-2019, 12:49 AM
I don't think that's necessarily a problem to explain; Bayesian skyline plots basically suggest to me that this is explained by larger increases / fewer population crashes in particularly East and South Asian population size between 50-10kya, which seems fairly consistent with those regions being warmer / less arid and having higher carrying capacity during the ice age.

See figures from Karmin 2015 (https://genome.cshlp.org/content/early/2015/03/13/gr.186684.114.full.pdf) - figures (https://imgur.com/a/ZJ59WdM)

It looks like autosomal dna isn't really subject to the same dynamics; subsequent greater increases in population size in South and Eastern Eurasia Asia and restricted population size in Western Eurasia don't do much to reverse the all conquering power of the serial OoA bottleneck.

Of course you can't take these totally literally, because for instance of course the South Asian mtdna pool was not actually necessarily evolving all in South Asia during that time period, and much of it must have subsequently come from West Eurasia, while the Siberian mtdna pool probably entered Siberia relatively late, etc. There are also more tumultous crashes and decreases and increases in reality that will be smoothed out in their curve. And y-dna is worse, more dislocated. But they are suggestive.

Think about the sheer number of primary branches there are of M, N, and R. Around 80. As it stands, not even 10 of them could plausibly be called West Eurasian. (For anyone curious, I'm counting the number of primary branches from Phylotree, and checking their distribution off Wikipedia).

I agree with the principle of serial bottlenecks from West to East reducing the heterozygosity of East Eurasians relative to the West. But that serial bottle-necking should also prune uniparental lineages out of the gene pool alongside autosomes. When populations crash and people die off, they take their whole genome with them. If 70/80 of the primary Eurasian mtDNA branches exist in East Eurasians today, and East Eurasians are essentially a subset of "West Eurasians" (don't get too caught up on the label) then implicitly that means West Eurasians must have had even more primary branches of uniparentals than what East Eurasians currently have, since the latter is a subset of the former (just like Eurasians vis-a-vis SSA). Yet somehow the West lost almost all of their primary branches through some sort of extreme bottleneck, but still maintained their greater autosomal heterozygosity over the East? It just doesn't add up to me.

The only plausible explanation I can come up with now is if the West Eurasian gene pool is significantly older than the East, with a subset of proto-Easterners diverging from the main Western group at a later time well after the Western group coalesced. Than the West goes through its own bottleneck which kills off many branches, but because it was much more deeply rooted it still maintains greater diversity in the autosomes. But in this case, I still think the surviving primary uniparental branches should show significantly older coalescence dates than what's in the East, which I don't believe is the case.

I'm not dogmatic about my position here, just very perplexed by what I'm seeing and am open to more discussion on this.

johen
03-06-2019, 06:14 PM
sorry.....

johen
03-06-2019, 06:18 PM
According to neolithic and broze achaeology of east asia, another type of people appeared: strange big ears, slanting big almond eyes, prominent nose and large mouth.

Hongshan neolithic:
https://1.bp.blogspot.com/-A4Cv06rDbA4/V2A3R-ivOfI/AAAAAAAAHm4/GVtAuiq_h3cIW3smjoJUZ-W2hgNUXX0egCLcB/s1600/Z-CHINA%2B013.jpg

Sichuan bronze: Yellow Di? Or not.
https://upload.wikimedia.org/wikipedia/commons/c/c2/Gold_Mask_%28%E9%BB%84%E9%87%91%E9%9D%A2%E7%BD%A9% 29.jpg

We can also meet this kind of large mouth and slanting big almond eyes in okunevo culture. Of course we can meet the eyes in maya and seima turbino culture. Actually seima trubino culture penetrated main land china and far reached to southeast asia 2,000bc. (maybe india also: https://anthrogenica.com/showthread.php?3433-Waves-of-migration-into-South-Asia&p=550417#post550417 (#3589)


These new observations suggest that the distribution of metal spearheads from the Seima-Turbino Culture to northern China represents the diffusion and spread of the metallurgical technique. From the metallurgical perspective in particular, the bronze casting of spearheads indicates the origin of piece-mould casting and core-casting technology, which influenced the bronze vessel casting method in China. We therefore suggest that the early Chinese metallurgy of the Lower Xiajiadian Culture in the western Liao River area can be linked to the Seima-Turbino Culture; this technique had spread from the Altai Mountain area to northern China via the Taosi Culture. After spreading to the Lower Xiajiadian Culture, it finally arrived at the Qijia Culture of Qinghai and Gansu provinces in the west

https://static.cambridge.org/resource/id/urn:cambridge.org:id:binary-alt:20170920130655-24645-optimisedImage-S0003598X17001776_fig1g.jpg
https://www.cambridge.org/core/journals/antiquity/article/origins-of-metallurgy-in-china/69E7A6EE63F38C7D5B2B0612E48DE264/core-reader

skyyrie
03-06-2019, 06:57 PM
Mt M seems gave birth to their most sublineage offspring in India.

And there is a OoA Out of Asia theory, which means a large number of people colonize the eurasian continent when the Sahara desert become too arid in Ice age.

There are papers about the appearance of Y haplogroup K2 and mt R in SouthEast Asia and Mt M in south Asia

Eterne
03-06-2019, 07:28 PM
@tuaman, I'm not sure what I'm thinking about totally adds up, but I'll try and explain.

I guess what I'm thinking is:

Say you have population A, and two descendents B, C. A has 3 mtdna dna types, B and C go through a bottleneck and retain 2.

C has gone through a harsher autosomal bottleneck initially but maintains both types, possibly through chance sampling, quickly goes to a higher population size, which allows it to maintain both types in the population and downstream mtdna to increase at a high rate.

B has gone through a weaker bottleneck, but maintains a low post bottleneck population size, meaning one mtdna type survives and few downstream clades, despite the population retaining more ancestral autosomal diversity from the weaker bottleneck.

This just kind of requires that mtdna and autosomal don't respond the same way to long term restricted population size and population increases vs bottlenecks and a different role for chance in fairly low size populations (autosome effected more by bottlenecks, less by drift in case of long term low pop size, reverse for mtdna). I'm not totally sure that this actually works, it would be good if somehow there was some modelling for this.

Alternatively, you could have some structured gene flow process like you describe, where the mtdna is actually coming from a restricted set of populations compared to paternal flow, enriching autosome relative to mtdna.

Or finally, there could just be an energetic or co-evolutionary consequence of M vs N - mtdna variation is much more likely to have selective signatures than y. See a papers on mito-nuclear co-evolution: https://www.biorxiv.org/content/10.1101/349126v1

tuaman, don't quite totally understand how your model works, probably my fault. could you do something visual to illustrate maybe? In terms of what we know from adna it doesn't seem like it's easy for the West Eurasian genepool to be substantially older than East Eurasian though - Tianyuan and Upper Paleolithic Europeans pretty much attested at the same time?

johen
03-07-2019, 05:25 PM
https://media.springernature.com/lw900/springer-static/image/art%3A10.1038%2Fs41598-018-35426-z/MediaObjects/41598_2018_35426_Fig3_HTML.png

If that migration in the research is correct, I think Sergei opinion would be persuasive. He thought that neolithic Yangshao people in the first culture of the yellow river were speaking altaic. Hg N was also found in the yangshao. Of course, there are counter arguments that the altai does not exist or Hg is not related to language. But his opinion is considerable under the migration, I think.

Moreover, modern Chinese is a tonal language, southern language, which appeared 4 century in China as I know (another migration from south?):

Tonal languages require humidity:
https://www.eurekalert.org/multimedia/pub/web/85653_web.jpg
https://www.eurekalert.org/pub_releases/2015-01/m-tlr012315.php

TuaMan
03-17-2019, 03:23 AM
Circling back to this, but regarding mtDNA M in particular with its 50+ primary branches, am I missing something but how exactly can a single uniparental have that many separate lineal descendants? Each primary branch derived from the tMRCA of M is supposed to represent one actual daughter of the founding mother of all M branches, so that would mean that tMRCA M woman had over 50 daughters?

Megalophias
03-21-2019, 06:40 PM
Circling back to this, but regarding mtDNA M in particular with its 50+ primary branches, am I missing something but how exactly can a single uniparental have that many separate lineal descendants? Each primary branch derived from the tMRCA of M is supposed to represent one actual daughter of the founding mother of all M branches, so that would mean that tMRCA M woman had over 50 daughters?
An mtDNA mutation only comes along on average about once in 2500 years. So you expect most of the maternal descendants of the M Matriarch will be carrying the same root haplotype for a long time afterward. Hence the M Matriarch only had to have 2 daughters, those branches can (and certainly do) represent many generations of branching, but they can't be distinguished because you don't have new mutations every generation.

In fact it is entirely possible that many primary M branches come from an expansion (or more than one) that occurred thousands of years after the initial M expansion, but with their ancestress being a woman who still carried the root haplotype of M, and hence is indistinguishable from the true root of M itself. Take haplogroup R, the main branch of N: it has 2 mutations under N - that lineage could by chance have had no mutations in that time, and all the branches of R would now be primary branches of N, even though the actual genealogy involved would be the same. IIRC this has actually been suggested as an explanation for why the TMRCA of M seems to be lower in India than elsewhere.

The same could have happened for N or L3 or anything else. E.g. there are 7 primary branches of L3, but they could be related in any way, e.g. M and N could really be one branch, or all the African branches could be one clade which is a sister to N, etc.

This is true of Y haplogroups as well, but mutations accumulate much faster, so it can only be a few generations blurred together.

ThirdTerm
03-29-2019, 09:48 PM
These findings from cranial measurements find extra support from non-metric dental morphology42, generally believed under strong genetic control and free of environmental influence, pointing to the same two layers of populations. One grouping is apparent in Australo-Papuan and early SEA teeth, consistent with the “first layer”. Another grouping is apparent in NEA and American natives, consistent with the “second layer.” Future research may consider the deeper relation between NEA and American populations, likely involving a shared ancestry through Siberia during the Pleistocene.


Matsumura et al. (2019) referenced recent papers in human genetics such as McColl et al. (2018) published in Science, which looked into the Two Layer model for the early peopling of Southeast Asia that is remarkably similar to the conclusion of this Craniometrics study. It would be impossible to come up with the Two Layer model by cranial measurements alone without relying on genetic studies. McColl et al. (2018) found that early genomes from Hoabinhian hunter-gatherers in Laos and Malaysia have genetic affinities with the Onge hunter-gatherers from the Andaman Islands, while Southeast Asian Neolithic farmers have a distinct East Asian genomic ancestry related to Austroasiatic populations. There were two further migratory events: 1) the expansion of speakers of Austronesian languages into Island Southeast Asia ca. 4 kya; 2) the expansion by East Asians into northern Vietnam ca. 2 kya.

http://science.sciencemag.org/content/sci/361/6397/88/F4.large.jpg



Ancient migrations in Southeast Asia
The past movements and peopling of Southeast Asia have been poorly represented in ancient DNA studies (see the Perspective by Bellwood). Lipson et al. generated sequences from people inhabiting Southeast Asia from about 1700 to 4100 years ago. Screening of more than a hundred individuals from five sites yielded ancient DNA from 18 individuals. Comparisons with present-day populations suggest two waves of mixing between resident populations. The first mix was between local hunter-gatherers and incoming farmers associated with the Neolithic spreading from South China. A second event resulted in an additional pulse of genetic material from China to Southeast Asia associated with a Bronze Age migration. McColl et al. sequenced 26 ancient genomes from Southeast Asia and Japan spanning from the late Neolithic to the Iron Age. They found that present-day populations are the result of mixing among four ancient populations, including multiple waves of genetic material from more northern East Asian populations.

Abstract
The human occupation history of Southeast Asia (SEA) remains heavily debated. Current evidence suggests that SEA was occupied by Hòabìnhian hunter-gatherers until ~4000 years ago, when farming economies developed and expanded, restricting foraging groups to remote habitats. Some argue that agricultural development was indigenous; others favor the “two-layer” hypothesis that posits a southward expansion of farmers giving rise to present-day Southeast Asian genetic diversity. By sequencing 26 ancient human genomes (25 from SEA, 1 Japanese Jōmon), we show that neither interpretation fits the complexity of Southeast Asian history: Both Hòabìnhian hunter-gatherers and East Asian farmers contributed to current Southeast Asian diversity, with further migrations affecting island SEA and Vietnam. Our results help resolve one of the long-standing controversies in Southeast Asian prehistory.

http://science.sciencemag.org/content/361/6397/88

TuaMan
03-30-2019, 01:22 PM
An mtDNA mutation only comes along on average about once in 2500 years. So you expect most of the maternal descendants of the M Matriarch will be carrying the same root haplotype for a long time afterward. Hence the M Matriarch only had to have 2 daughters, those branches can (and certainly do) represent many generations of branching, but they can't be distinguished because you don't have new mutations every generation.

In fact it is entirely possible that many primary M branches come from an expansion (or more than one) that occurred thousands of years after the initial M expansion, but with their ancestress being a woman who still carried the root haplotype of M, and hence is indistinguishable from the true root of M itself. Take haplogroup R, the main branch of N: it has 2 mutations under N - that lineage could by chance have had no mutations in that time, and all the branches of R would now be primary branches of N, even though the actual genealogy involved would be the same. IIRC this has actually been suggested as an explanation for why the TMRCA of M seems to be lower in India than elsewhere.

OK, the much slower mtDNA mutation rate makes sense of how there can be so many independent primary branches. I guess I'm still a bit confused on why one would think that it has an impact on the calculation of the tMRCA of M in India relative to other M clades in Asia. Why not just go with the more straighforward assumption that M is simply younger in India than East Asia? I mean there are multiple branches of M in India, what are the odds that all of them just had no mutations at all for several thousand years while the East Asian branches mutated along at the normal pace?

Megalophias
03-30-2019, 04:56 PM
I mean there are multiple branches of M in India, what are the odds that all of them just had no mutations at all for several thousand years while the East Asian branches mutated along at the normal pace?
The idea isn't that a whole bunch of branches had no mutations, it's that one branch which happened to not have any mutations expanded into many daughter branches. If that's what happened, it's only a possibility.

ThirdTerm
03-30-2019, 06:35 PM
I guess I'm still a bit confused on why one would think that it has an impact on the calculation of the tMRCA of M in India relative to other M clades in Asia. Why not just go with the more straighforward assumption that M is simply younger in India than East Asia?

The founder age of M in India (39.70 ± 3.24 ky) is significantly younger than those in East and Southeast Asia (55.60 ± 2.94 ky). The carriers of the M lineages spread from southeastern Asia, reaching India westwards and near Oceania eastwards.



As the most parsimonious topology, the existence of several M superhaplogroups, embracing different lineages based on one or a few moderately recurrent mutations have been recognized in PhyloTree.org Build 16 [44]. Following this trend we have, provisionally, unified haplogroup M11 and the recently defined haplogroup M82 [92] under macrohaplogroup M11′82 because both share the very conservative transition 8108 (Additional file 2: Table S2). Attending to their phylogeography these superhaplogroups usually link very wide geographic areas (Additional file 2: Table S2). In accordance with our suggestion that the carriers of Macrohaplogroup M colonized South Asia later than southeastern Asia and Oceania and that this mtDNA gene flow had an eastern origin we have observed that the mean radiation age of those superhaplogroups involving South Asia (39.70 ± 3.24 ky) is significantly younger (p = 0.003) than the one relating East and Southeast Asia (55.60 ± 2.94 ky). In this comparison we considered the South Asia-Southeast Asia-Australian link deduced from superhaplogroup M42′74 as specifically involving India. However, we considered superhaplogroup M62′68 as indicative of an East Asia-Southeast Asia link because M62 has been found consistently in Tibet [112, 113] but only sporadically in northeast India [67].
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5105315/

palamede
04-02-2019, 09:35 AM
The founder age of M in India (39.70 ± 3.24 ky) is significantly younger than those in East and Southeast Asia (55.60 ± 2.94 ky). The carriers of the M lineages spread from southeastern Asia, reaching India westwards and near Oceania eastwards.

I suppose the Out of Africa isthe oldest hypothesis during Eemian dating around 130-120,000 coinciding with the Nubian Culture in Arabia and thed AMH remains in Levant (y-hg CT and mt hg L3'4'6) . it could be an out of Arabia/Into East Asia) happened during the more temperate (with warm peaks) and wetter Odderade phase (85,000-75,000) before the first pleniglacial (75,000-60,000). During this expansion to India, South China and Sunda of y-hg C, D, H, K and mt hg M and N, there was a star explosion of hundred of M sub-branches.
But why the founder age of M branches India seems younger the founder ages of M in South-East Asia although I think the real foundation date was the same and probably some centuries earlier in India : one reason is the big nowadays M branches in India (M2, M3, M4'30, M5, M6) plus the occidental M1 which seem me indigenous and not coming from East Asia. They are a big part (60%) of the Indian population and rare elsewhere.

1) A secondary reason : a new extension in Sahul (Australia/New Guinea and emerged platform during the first pleniglacial which generates new branches M branches an sub-branches which have survived until nowadays.

2) A main reason with two coupled reasons : the cataclysmic volcanic explosion of Toba in Noth-West Sumatra around 74-73,000BP and the first pleniglacial (75-60,000) with great extension of deserts and poor vegetation which affected a quick population reprisal after the catastroph of Toba Volcano https://en.wikipedia.org/wiki/Lake_Toba . It seems the dominant winds created big ash layer specially in India, this should have killed the greatest part of population of India but probably in sheltered areas there were some survivors bearing Y haplogroup H and mt hgs specifically indian ; a lot of branches and sub-branches of M disappeared in India who created the artifiicial younger indian M.

There was some migrations coming from East to North India in the emptiness created by the catastroph, but not a lot because the coming first pleniglacial . It could be the time Y-hg P and several mt branches of R come from Sunda/Indochina to North India and then norhwards in Central Asia/Siberia after the end of pleniglacial around 60000BP.

The advantages and the problems of this hypothesis is for its compatible mutation rates to facilitate the explanation of other genetic events , but also the difficulties to conciliate other genetic events as well as to question the nowadays consensus-accepted mutation rates. I am not a specialist, but maybe there are differences between slower evolutionary rates and real biological rates.

https://en.wikipedia.org/wiki/Haplogroup_M_(mtDNA)