And also, we need to separate the Basal .. There are early Archaic Basal, Early-Middle , Middle, Late-Middle , Late ... Or we may find some better Time definition for the Basals.
We have also Basal in North-East Asia, some in South_East Asia, there is Basal in America too.
In my opinion the very first OOA to East Asia are 100% Basal only.. () Why we can be sure about it ?
Because if they were mixed, then we would see different proportions.
However there is not that much Basal in Papua NG and Australia, there is some, but it would come later. The Papuans were not Basal.
Quote:Early Upper Paleolithic genomes of Crimea show migration and admixture dynamics of the first modern European ancestries
Populations migrated from the Middle East through the Caucasus since the beginning of the Pleistocene. Our analysis of the genomes of 37-36,000 year-old skeletal remains from Buran Kaya III in Crimea revealed that these individuals were among the earliest to migrate into Europe after a major climatic crisis, the Heinrich Event 4 and the Campanian Ignimbrite volcanic super-eruption around 40,000 years ago. Their genomes also contained traces of the genomes of the earliest sapiens populations living in Europe and associated with the Initial Upper Paleolithic. This demonstrated that the newer migrants did not simply replace the earlier inhabitants of Europe, but some intermixing between the groups did occur. The genomes of the individuals from Buran Kaya III further revealed strong genetic relationships with pre- and post-Glacial populations from the Caucasus in agreement with archaeological evidence pointing to extended social contacts from the Zagros to the Carpathian mountains...
So is anyone going to have an attempt at deep East Eurasian phylogeny, should we have a separate thread for that?
Shahr-I-Sokhta BA2 is a straightforward mix with some kind of AASI, as are a lot of modern Indians, we have all kinds of Oceanians, Andamanese, Jomon, Amur River, Chinese ancients, one of the Hoabinhians is good quality, maybe Longlin can be made to work too.
That IUP type C1b that appeared in Vestonice is a good example of that admix that took place between IUP and the newcomers from the Caucasus
We can't talk about deep West Eurasian phylogeny without discussing the question about "Basal".
It is almost "forbidden" word in this forum..
"Basal" is much more complicated and vefy unclear, doesn't matter how many years of experiance you may have.
May we discuss such question here or we need to open a new topic?
04-11-2024, 11:16 PM (This post was last modified: 04-11-2024, 11:32 PM by Kale.)
Bit of an update, a few minor tweaks to improve the graph I posted on page 17, post 251 "121123f". Worst F3 residual was 1.8
Again since these graphs are huge I'll give a pop-by-pop overview.
General deep structure: ANA splits first (X0 node), Basal splits second (X2), crown Eurasians are X3. 'X' is an archaic node for BachoKiro
- ZlatyKun: Basal + generic crown
- Ust-Ishim: falls gently on the Western side of Crown Eurasians
- BachoKiro: falls gently on the Eastern side of Crown Eurasians (plus some archaic)
- East Eurasians split into 'North' (Tianyuan) and 'South' (Onge)
- Goyet: The archetypal West Eurasian
- 'EEE' (Early East European): 84/16 mix of Goyet/Basal. Splits into a Kostenki/Sunghir branch and a proto-CWE branch.
- ANE: 76/24 mix of proto-Kostenki/Sunghir and Southern ENA. Yana gets 7% Northern ENA also.
- East-Asians (Jomon, Primorsky_N): 73/27 mix of Soutern-ENA and what effectively stands in for Salkhit (even mix of Northern-ENA and Yana)
- Native Americans: 73/27 mix of Primorsky_N/ANE
- BK1653: 61/39 mix of proto-CWE and Goyet
- Muierii: 84/16 mix of BK1653/Sunghir
- Gravettians (Vestonice, KremsWachtberg, Ostuni, Paglicci): 53/47 mix of BK1653/Sunghir
- WHG (Italian Epigravettians): 70/24/6 mix of CWE/Gravettian/ANE
- Georgia_UP: 73/27 mix of CWE and the basal that contributed to EEE.
- 'Zarzian': 71/29 mix Basal/Southern-ENA
- Pinarbasi: 70/30 mix CWE/Zarzian
- IBM (Taforalt): 65/35 mix of ANA and Pinarbasi-related CWE
- NE2 (I don't know what to call this one): 71/29 mix CWE/ANE
- Iran_N: 69/31 mix Zarzian/NE2
- CHG: 52/48 mix Zarzian/NE2
- EHG: WHG/ANE mix obviously, but with 25% Pinarbasi and 7% Primorsky_N
Keep in mind there are 2 equal ways of accounting for basal. Either the way depicted here (ZlatyKun = Basal + generic Crown) or the way depicted previously where ZlatyKun is pure basal, and Near-East basal is a fusion of that and 'super-basal'
Personally I didn't like the idea of CWE for a long time, but maybe that's just because scholars were touting WHG as somehow being preserved relics of 100% pure CWE.
I really do like this sort of 'soft-CWE' model here, where CWE is just a regular paleo-Euro descended group like Kostenki, and where WHG is allowed to be mostly CWE, but also have a decent amount of Gravettian ancestry and low-level ANE.
If we use this diagram as an example, the West Eurasians are the branch which separated from Basal Eurasians earlier.
So let try to specify the definition for West Eurasians ..
1. We have OOA - The most significant recent dispersal of modern humans from Africa gave rise to an undifferentiated "non-African" lineage by some 70–50 ka (70-50,000 years ago).
2. OOA are separated on 2 directons: Basal + Non-Basal Basal are those who stay in Near East ( NE) Non-Basal are those who continue the migration in other directions.
3. Types of non-Basal -3 .1 Papua/ Australia/ New Guinea - PNG -3. 2 East Asians - EA -3. 3 West Eurasians - WEuA
-3.4 There should be also a North direction: for Central Asia and North Eurasians ( CA + NEuA )
These groups are under many conditions: because any of them PNG, EA , WEuA may also have some "Basal". More than that: by definition and by origin all 3 of them are "Basal". The question will be how they become non-Basal ?
There was some admixture with "External" groups, separation by time and as a result of such admixtures these former "Basal" will become non-Basal. And here we should start the discussions about such "external" , archaic, unknown hominoids that contributed for EA, WEuA and PNG.
The lowest admixture with external archaic seems to happen in the Northern direction and that's why we register the highest amount of "Basal" in the northern groups like Iran and CHG.
04-26-2024, 11:03 AM (This post was last modified: 04-26-2024, 11:03 AM by Jerome.)
(04-12-2024, 01:45 PM)TanTin Wrote:
If we use this diagram as an example, the West Eurasians are the branch which separated from Basal Eurasians earlier.
So let try to specify the definition for West Eurasians ..
1. We have OOA - The most significant recent dispersal of modern humans from Africa gave rise to an undifferentiated "non-African" lineage by some 70–50 ka (70-50,000 years ago).
2. OOA are separated on 2 directons: Basal + Non-Basal Basal are those who stay in Near East ( NE) Non-Basal are those who continue the migration in other directions.
3. Types of non-Basal -3 .1 Papua/ Australia/ New Guinea - PNG -3. 2 East Asians - EA -3. 3 West Eurasians - WEuA
-3.4 There should be also a North direction: for Central Asia and North Eurasians ( CA + NEuA )
These groups are under many conditions: because any of them PNG, EA , WEuA may also have some "Basal". More than that: by definition and by origin all 3 of them are "Basal". The question will be how they become non-Basal ?
There was some admixture with "External" groups, separation by time and as a result of such admixtures these former "Basal" will become non-Basal. And here we should start the discussions about such "external" , archaic, unknown hominoids that contributed for EA, WEuA and PNG.
The lowest admixture with external archaic seems to happen in the Northern direction and that's why we register the highest amount of "Basal" in the northern groups like Iran and CHG.
Do you know how Neanderthal estimation is done using fstats?
Lazaridis mentioned something about F4 ratios but is there any other method.
(04-26-2024, 11:03 AM)Jerome Wrote: Do you know how Neanderthal estimation is done using fstats?
Lazaridis mentioned something about F4 ratios but is there any other method.
They do some estimation directly from F-stats. I have some idea and I have some doubts about it.
It depends on too many things. There are certain regions where we may apply such calculation. And there are other regions where we can't use such methods.
The best method is when researchers are checking directly in the raw data and searching for regions with long consecutive snips to compare that with Neanderhtals.
If we calculate by single snips: the origin of such snips could be also from other archaic spieces like Denisova or Gorilla. Same snips could be present in other spieces.
When you compare long consecutive range of snips - then we have high certainity these originated from Neanderthals and not from something else.
(04-26-2024, 11:03 AM)Jerome Wrote: Do you know how Neanderthal estimation is done using fstats?
Lazaridis mentioned something about F4 ratios but is there any other method.
I will give you an example with our test for Basal/Neanderthal that we did earlier.
"F4 ( Chimp.REF Vindija_Neanderthal.DG ; Yoruba.DG TEST )"
(04-26-2024, 11:03 AM)Jerome Wrote: Do you know how Neanderthal estimation is done using fstats?
Lazaridis mentioned something about F4 ratios but is there any other method.
I will give you an example with our test for Basal/Neanderthal that we did earlier.
"F4 ( Chimp.REF Vindija_Neanderthal.DG ; Yoruba.DG TEST )"
Now let divide Fstats with "F4 ( Chimp.REF Vindija_Neanderthal.DG ; Yoruba.DG Vindija )" and will multiply by 100 to get the percentage.
- This is our fstat for Vindija which is the maximum.
So we do this ratio:
F4 ( Chimp.REF Vindija_Neanderthal.DG ; Yoruba.DG TEST ) / F4 ( Chimp.REF Vindija_Neanderthal.DG ; Yoruba.DG Vindija ) *100
Here is the full picture: 1-fstat 2 - percentage
Thanks,one minor request.
Could you also try adding Sunghir, Kostenki14,Vestonice16,Krems WA1 and also some of pre-WHG post-LGM Europeans like el miron?
(03-14-2024, 12:51 AM)TanTin Wrote: We can't talk about deep West Eurasian phylogeny without discussing the question about "Basal".
It is almost "forbidden" word in this forum..
"Basal" is much more complicated and vefy unclear, doesn't matter how many years of experiance you may have.
May we discuss such question here or we need to open a new topic?
Do you think Basal eurasians were kind of an in-group African lineage responsible for most of the non archaic ancestry in Africans?
Tossing another idea around, as I said in my last post here... "there are 2 equal ways of accounting for basal.
1) Either the way depicted here (ZlatyKun = Basal + generic Crown)
2) ZlatyKun is pure basal, and Near-East basal is a fusion of that and 'super-basal'".
Those both didn't fully satisfy me from a geographic perspective because...
1) Presumably X3 ('Crown Eurasians') formed in Central Asia, given the two primary branches are Ust-Ishim and East Eurasians. Already I have Goyet and co. siding on the Ust side of things, so that's 1 Westward migration, ZlatyKun getting it's own X3 branch poses a second Westward migration. Doesn't sound too elegant.
2) Where and when would 'super-basal' have been hiding that it could A) Not be in ZlatyKun, B) Not be in the X3 node, C) be in Zagros by ~40kbp to mix into Kostenki and co.
In both cases the specific interaction between ZlatyKun and Basal Eurasians (The X2 > X2a drift as per graph 41124b) struck me as a bit odd, as it leaves very little room for Crown Eurasians to have left any sort of genetic trail in Zagros/Iran.
So here's a third option.
Stage 1) A (late Middle Paleolithic?) migration from Levant to Zagros/Iran, staying put for the time being ('Super-basal' in my most recent modeling)
Stage 2) A second (IUP?) expansion from Levant, 1 way goes North to Europe as ZlatyKun, another to Zagros/Iran again, mixing with the first wave / super-basal, and spreads further becoming Crown Eurasians in Central-Asia.
The only problem at the moment is that, since everybody (except ZlatyKun) has this Zagrosian-super Basal in equal quantities, there is nothing to constrain when it branched off, or what % it contributed. For that I'll have to wait until Ranis comes out. Consequently I put super-basal coming off at the same point as ANA, but don't take that to mean anything.
As for the migration route it becomes more straightforward. X0 (again ignoring it's specific placement) would be Levant-wave1, X1 is Levant-wave2, X3/X4 are Zagros, X4a/X4b stay put in Zagros, X5 is Central Asia = Crown Eurasians.
04-30-2024, 07:21 PM (This post was last modified: 05-01-2024, 02:20 AM by Kale.)
(04-29-2024, 02:31 PM)Kale Wrote: The only problem at the moment is that, since everybody (except ZlatyKun) has this Zagrosian-super Basal in equal quantities, there is nothing to constrain when it branched off, or what % it contributed. For that I'll have to wait until Ranis comes out. Consequently I put super-basal coming off at the same point as ANA, but don't take that to mean anything.
I have to correct myself, there is one major constraint I overlooked, which is that the members of the X3/X4 nodes (Zagros after the mix), would be more a good bit more related to ZlatyKun than to each other. Reviewing the residuals I noticed the admix % was actually quite tight, so the drift edges should have some validity. But a 67 unit drift edge for Wave2 (ZlatyKun), versus an 8 unit drift edge (maximium) for wave1, is quite the imbalance, not to mention the scale of a 67 unit edge, which is bordering on implausibly strong.
So here's a bit of a correction. The ZlatyKun drift edge is tamed to 38 units, which is still rather high, but at least within the realm of plausibility (the ANE proper bottleneck is 49 units). Again, Ranis will be crucial to determine whether this is actually valid. Noticeably ANA actually sits on the Wave1 (X2) branch, which would imply the OOA bottleneck to be rather soft and maybe that ANA formed in the Levant and backmigrated to Africa. I might try and stick Mota in there when I get some time to make sure it's not too soft.
EDIT: Added Mota to the graph, X0/'OOA' has 11 units of drift post-Mota.
![[Image: Basal-Eurasian.png]](https://i.postimg.cc/855mvmmJ/Basal-Eurasian.png)
