Zeng et al: Postglacial genomes from foragers across Northern Eurasia...

[Jaska]

Kale:

“Yes it seems Jaska's view on the genetic evidence is often misinterpreted, so I'd like to offer a comparison of counterparts with the more familiar PIE in hopes it may be helpful.
PIE ~ Pre-proto-Uralic
Yamnaya ~ Yakutia_LNBA/Kra001
Indo-Iranian ~ proto-Uralic
Sintashta ~ Mostly Yakutia_LNBA/Kra001, but with some other stuff

So to say Proto-Uralic is associated with Yakutia_LNBA/Kra001 ancestry is the equivalent to saying Indo-Iranian is linked to Yamnaya ancestry.
That's not technically wrong, given Steppe_MLBA are ~2/3 Yamnaya-related, but clearly linking Indo-Iranian to Steppe_MLBA is more correct.”

And how do you know that those people with Steppe_MLBA ancestry were speakers of Late Proto-Indo-Iranian? Because they happen to be in the right place at the right time: within the Sintashta Culture, which contains the most accurate material matches for the LPIIr vocabulary concerning chariots and certain ritual concepts described in Rigveda and Avesta.

With Uralic languages you must follow exactly the same procedure: you must take into account all the relevant linguistic arguments for dating and locating Late Proto-Uralic. Only after that you can see, which kind of population was present at the right place at the right time.

However, many people here do not follow the scientific procedure. Instead, they just arbitrarily decide which ancestry is related to Proto-Uralic speakers. It is easy to illustrate the absurdity and unreliability of such a method:

1. The majority ancestry in Nganasans is the Yakutia_LNBA ancestry. Nganasans speak a Uralic language. Therefore, Uralic language is connected to the Yakutia_LNBA ancestry.

2. The majority ancestry in Finns is the Corded Ware ancestry. Finns speak a Uralic language. Therefore, Uralic language is connected to the Corded Ware ancestry.

If we follow all the ancestries in Uralic populations back in time, we get many competing associations for the Uralic language. From DNA it is not possible to see, which answer of those is true, but only one can be – all the other associations are false. The only way to find the right answer is to take the linguistic results and find a match for them.

People also erroneously believe that one ancestry is related to only one language. Therefore their follow-up mistake is to believe that language can be seen from DNA: if certain ancestry in one area at one time is connected to certain language, they believe that the same ancestry in another area at another time testifies for the same language. That is not true, because the presupposition for it is not true.

Now, if you or somebody else want to claim that the Uralic language derives from the population like Yakutia_LNBA (or whatever ancestry, for that matter), you are welcome to present your evidence here. Show me that you follow the scientific procedure and are not just pulling one out of your hat.

[Queequeg]

I know some argue they are recent language shifters or so but i honestly have not seen any convincing evidence for this.

According to Gill et al 2023 Nganasans seem to be based on the same Transbaikal_MN and Yakutia_MN type of a mixture as fex kra001. Nganasans however are somewhat more biased towards Yakutia_MN i.e. they have more northern, ANE based elements. This might possibly point to some level of language switch into Uralic, provided of course that at least some group within Yakutia_LN already spoke Uralic. Yakutia_LN, apparently getting close to Yakutia_LNBA of Zeng et al, is on the other hand more biased towards East i.e. Amur, which probably explains why according to f3 one of the closest ancient matches of Yakutia_LNBA is Xianbei_IA, despite the fact that the group is also based on northern Yakutia_MN group.

Nganasan = Yakutia_MN 59,5% + EastBaikal_N 40,5%
Krasnoyarsk_BA (kra001) = Yakutia_MN 47,5% + EastBaikal_N 52,5%

Yakutia_MN = Dzhylinda-1 (irk00x) 73% + WestBaikal_EN 27%
Yakutia_LN = Yakutia_MN 45% + EastBaikal_N 55%

[Jaska]

Andar:

“I know some argue they [Nganasans] are recent language shifters or so but i honestly have not seen any convincing evidence for this.”

Here is the evidence:
1. Nganasans are the most northern Samoyedic population and therefore most distant from the Samoyedic homeland.
2. Nganasans differ from all the other Samoyedic populations: the others have much more complex genetic composition.
3. Tundra Yukaghirs, earlier spoken much closer to Taimyr, have even more Yakutia_LNBA ancestry than Nganasans.

It would be incredible to assume that all the other Samoyeds changed their genetic composition, while Nganasans marched the longest journey without changing. More credible is to assume that Nganasan language carriers admixed heavily with Paleo-Taimyrians, which resembled Tundra Yukaghirs.

[Kale]

Now, if you or somebody else want to claim that the Uralic language derives from the population like Yakutia_LNBA (or whatever ancestry, for that matter), you are welcome to present your evidence here. Show me that you follow the scientific procedure and are not just pulling one out of your hat.

I hope you don't think I was trying to argue with you. I was trying to help convey your point.

Uralo-Yukaghir relatedness is still far from proven, and there are other interesting propositions, like Uralo-Eskimo relatedness. Perhaps this could also agree with the Yakutia ancestry?

There are actually multiple genetic layers linking Uralic and Eskimo groups...
The timeline of genetics in Yakutia goes something like...
Mesolithic: Kolyma_meso (similar to New World)
Neolithic: ~50% influx of something most similar to Baikal_EN > result = Yakutia_MN (N5a.SG) & Saqqaq (they form a clade)
LN/BA: ~50% Transbaikal_N ancestry arrives > result = Yakutia_LNBA/Kra001

The Neolithic stage deserves some elaboration because there is important structure within the Yakutia_MN/Saqqaq clade.
- The type of ancestry in Yakutia_LNBA is generic, not preferring one or the other.  
- A generic or mildly Saqqaq related group reached British Columbia Canada as early as 4500BC. 
- The type in Athabaskans is perhaps a bit more Saqqaq leaning, or maybe it's just closer to significance because they have more than the Earlier Canadian samples.
- Magadan_BA is about an even mix of Kolyma_Meso and a strongly Saqqaq related source.

Quick runs on some modern populations from NE-Siberia.

Eskimo-Aleut:
Eskimos (Russian ones from Chaplin, Naukan, and Sireniki): ~45% Saqqaq, 35-40% New World, 15-20% Yakutia_LNBA
Russia_Ekven_IA.SG: Same as above.
USA_AK_NeoAleut: ~45% Saqqaq, ~50% New World, Yakutia_LNBA 5% (but standard error overlaps 0)

Chukotko-Kamchatkan: 
Itelmen - Mostly Magadan_BA, good chunk of Yakutia_LNBA. 
Chukchi - same as above, but with a large amount of European ancestry also.
Koryak - couldn't get a passing model

Tungusic: 
Ulchi: 2/3 similar (but not exactly) DevilsCave, tiny bit of Jomon, ~15% Magadan_BA, ~15% Yakutia_LNBA
Even: ~2/3 Yakutia_LNBA, ~1/4 DevilsCave-like, maybe some other stuff, couldn't quite nail it down.

So if proto-Uralo-Eskimo was a thing, what did the genetics of that group looked like? There's a lot of options. 
I do find the genetic transmission chain of Baikal_EN > Yakutia_MN/Saqqaq > 4500BC Canada & Athabaskans intruiging for Dene-Yeneseian though.

[Psynome]

I do find the genetic transmission chain of Baikal_EN > Yakutia_MN/Saqqaq > 4500BC Canada & Athabaskans intruiging for Dene-Yeneseian though.

Yes, I believe Dene-Yeniseian is on stronger footing than Uralo-Eskimo, from a genomic, linguistic, and historical/archaeological standpoint. 

Linguistically, the shared lexicon between Uralic and Eskimo-Aleut seems to me much weaker than that between Uralic and Yukaghir. The Uralic Yukaghir correspondences have been dismissed by multiple linguists as loans, yet they occur heavily in basic high use words and especially verbs which are typically the least susceptible to replacement. Rather, to me it appears more as evidence of a shared core vocabulary that persisted as more specialized lexemes were replaced.

Yakutia_LNBA is a simple genomic link between the two populations, and considering the massive size of the Late Neolithic Ymmyyakhtakh horizon, it's not hard to place the ancestors of both Uralic and Yukaghir within it.

[Jaska]

Kale:

“I hope you don't think I was trying to argue with you. I was trying to help convey your point.”

OK, in that case it seems that I confused subject and object of the word “misinterpreted” in the first sentence of your earlier comment.

Psynome:

“Linguistically, the shared lexicon between Uralic and Eskimo-Aleut seems to me much weaker than that between Uralic and Yukaghir. The Uralic Yukaghir correspondences have been dismissed by multiple linguists as loans, yet they occur heavily in basic high use words and especially verbs which are typically the least susceptible to replacement. Rather, to me it appears more as evidence of a shared core vocabulary that persisted as more specialized lexemes were replaced.”

Nowadays we know that everything can be borrowed, if the contact is intense enough. Grammatically and lexically Yukaghir shares more with Samoyedic than with Proto-Uralic; this is one reason why many features and words are considered borrowings. Still, borrowing does not exclude relatedness, but so far the evidence for relatedness has not been sufficient – but not for any other language family, either. Uralo-Yukaghir is no more probable than Indo-Uralic, and they both cannot be true at the same time. One pulls to the east, another to the west.

Psynome:

“Yakutia_LNBA is a simple genomic link between the two populations, and considering the massive size of the Late Neolithic Ymmyyakhtakh horizon, it's not hard to place the ancestors of both Uralic and Yukaghir within it.”

Ancestors for Uralic and Yukaghir populations, sure – but not necessarily linguistic ancestors. Although for Yukaghir that seems possible, but as long as we do not know did Pre-Proto-Uralic originate in Europe or Siberia, we have no basis to connect the Uralic language lineage to Yakutia_LNBA ancestry.

There are same ancestries in populations belonging to different language families, and there are different ancestries in populations belonging to the same language family. In one place at one time an ancestry may be connected to one language, in another place at another time to another language.

There are tens of expansions from the east to the west and to the west to the east, and every one of them was connected to some language. Even when some ancestries have wide distribution in the Uralic populations, it does not mean that they are connected to the spread of the Uralic language. Some ancestry may be too early, some other too late – only an ancestry which happens to be in the right place at the right time can be connected to the spread of the Uralic language. And only linguistic results can give those coordinates to which to compare.