PDA

View Full Version : Modern distribution of R1b-Z2103



Maciamo
12-02-15, 19:27
The new Haak et al. 2015 (http://biorxiv.org/content/biorxiv/early/2015/02/10/013433.full.pdf) paper confirmed that Yamna Proto-Indo-Europeans belonged to haplogroup R1b. Four out of six R1b samples from the Volga-Ural region belonged to the R1b-Z2103 subclade, a branch of what used to be called R1b-ht35, the eastern variant of R1b-M269. Obviously the samples tested were on the far north-eastern reaches of the Yamna horizon, and I expect that samples from more southern and western areas of Yamna would yield other R1b subclades, notably the L51 branch from which Western Europeans descend.

We can reasonably assume that all R1b-M269 samples that once fit into the ht35 or L23* category are Z2103, simply because there are only two known subclade under L23 : L51 (ht15) and Z2103 (ht35).

http://cdn.eupedia.com/images/content/Haplogroup-R1b-Z2103.png

What is not clear from my European map above is that Z2103 has a very wide distribution covering also Central Asia, South Asia and West Asia.

Most of the R1b in Asia on the map below is either M73 or Z2103, although the latter is dominant everywhere.

http://cache.eupedia.com/images/content/Haplogroup_R1b_World.png

According to the R1b1a2 (P312- U106-) DNA Project (https://www.familytreedna.com/public/ht35new/default.aspx?section=yresults), Z2103 has five main subclades.

- L584 (including L943) : found mostly in the South Caucasus, Turkey, Iraq and Iran, but also in Poland, Germany and Austria.

- L277.1 : found in Russia, Central Asia, Bulgaria, India and the Middle East (Iraq, Lebanon).

- CTS7822 (including CTS9219) : found in Russia (including Chuvashia), Ukraine, Poland, Hungary, the Balkans, Armenia, Italy, Spain, Britain and Ireland.

- CTS7763 : found in Turkey, Bulgaria and Italy.

- Z2109 : found in Germany and Scotland.


Overall the distribution of the bulk of Z2103 samples is reminiscent of that of R1a-Z93.

For example R1b-L277.1 seems to have expanded from Russia to Central Asia then to India and the Middle East, just like the R1a-L657 subclade of Z93.

R1b-L584 looks more Mitanni, Iranian, Scythian and maybe also Armenian. It's West Asian distribution matches that of the R1a-Z2124 subclade of Z93.

R1b-CTS7822 is mostly central and eastern European and correlates more with R1a-Z280.

R1b-CTS7763 appears to be confined to the greater ancient Greece (not data from Greece itself, but most of the Greek R1b-L23 could belong to this subclade considering its presence in South Italy and West Anatolia). It doesn't seem to correlate with any R1a subclade.


I wouldn't be surprised if the eastern Yamna Z2103 samples tested later formed the Poltavka culture (http://en.wikipedia.org/wiki/Poltavka_culture), which eventually merged with the Abashevo culture (http://en.wikipedia.org/wiki/Abashevo_culture) to form the Sintashta culture (presumably the main source of Asian R1a-Z93). Through a founder effect or through political domination, R1a-Z93 lineages would have become far more numerous than R1b-Z2103 after the expansion to Central and South Asia.

What is certain is that one of the Z2103 samples came from the Orenburg oblast just south of the Urals, in what was soon to become the Poltavka culture. I already envisioned this scenario in my migrations maps 5 years ago, and the current data is in perfect agreement with it. Back then Z2103 had not yet been discovered, so I couldn't have made predictions about it. Even today data is very scarce about L23 subclades, especially among Volga-Ural ethnic groups, in Central Asia and in the Balkans.

http://cdn.eupedia.com/images/content/early_middle_bronze_europe.png

Finalise
12-02-15, 20:07
Why has no one mentioned Bashkirs? They're very high in RM269, and are probably Z2013. They also live in today's Samara.This shows that western Yamna would probably be similar to what Ukraine is like today.

Robert6
12-02-15, 21:37
R1b-Z2105 in Volga region (according to Trofimova 2015)
36.2% Burzyan Bashkirs
21.2% Udmurts
8.0% Komi
6.8% Erzya and Moksha
3.8% Besermyan
2.3% Chuvash
0% Mari
0% Kazan Tatars,
0% Bashkirian Tatars,

bicicleur
13-02-15, 00:07
very interesting

it must have been a very mobile tribe

what about Armenians / Phrygians ?
on the move and entering history 1200 BC or a bit later

Fire Haired14
13-02-15, 01:30
R1b-Z2105 in Volga region (according to Trofimova 2015)
36.2% Burzyan Bashkirs
21.2% Udmurts
8.0% Komi
6.8% Erzya and Moksha
3.8% Besermyan
2.3% Chuvash
0% Mari
0% Kazan Tatars,
0% Bashkirian Tatars,

Thanks for that info. So, it's as high in parts of Russia as in west Asia. Now the R1b-Z2103 in Samara Yamnas makes more sense.

As of far 3/3 Corded ware have R1a1-M417, 3/3 Bell beaker have R1b(One was tested for P312 and was positive), all but a few of the dozens of Y DNA samples from Bronze-Iron age IE north Asians have R1a, 2/2 Mesolithic Russians have R1b1* and R1a1*, and Upper Palaeolithic MA-1 has R*.

That's an incredible continuum and dominance of Y DNA R in pre-historic ANE-heavy pops of Europe and north Asia. None of the Yamna R1bs were related, all their paternal lineages were not connected recently. This wasn't the result of a recent founder effect, Yamna legitally had a very high amount of R1b.

Who's to say R1b-L11 in west Europe is not descended of the steppe? It's pretty obvious now that we know Yamna and modern eastern Russians have a high amount of L51's brother Z2013.

Fire Haired14
13-02-15, 01:35
Let's say R1b-L23 originated and spread out of modern Russia.

It's debatable where R1b-M269 originated, but M269's brother M73 as far as know is centered in central and north Asia. This supports a north Eurasian origin of R1b-P297. Other R1b could have orignated somewhere else.

As of far we already have R1b1* from Mesolithic Samara and Neolithic Spain ancestral to P297. The R1b1* from Spain could be coming out of Russia or west Asia. But considering R1a1* has been found in Mesolithic Karelia, and most modern R1a is decended out of the same area most modern R1b is, I tend to think R1 in general might be from north Eurasia.

Maciamo
13-02-15, 09:10
Why has no one mentioned Bashkirs? They're very high in RM269, and are probably Z2013. They also live in today's Samara.This shows that western Yamna would probably be similar to what Ukraine is like today.

Why do you say "no one" when I am the only person who had posted in this thread until then ?

As I said above there is very little data on L23 subclades from the Volga-Ural region at the moment. I checked all the FTDNA projects (Bashkir Suyun, Bashkorostan, Chuvash, Tatars, Mordovians) and only found deep L23 subclades tested for the Chuvash. As I explained, there is a good chance that most if not all L23 in the region is Z2103, as well as in Central and South Asia.

Maciamo
13-02-15, 09:12
R1b-Z2105 in Volga region (according to Trofimova 2015)
36.2% Burzyan Bashkirs
21.2% Udmurts
8.0% Komi
6.8% Erzya and Moksha
3.8% Besermyan
2.3% Chuvash
0% Mari
0% Kazan Tatars,
0% Bashkirian Tatars,

Thanks, I hadn't seen this study published only a few weeks ago. Very timely.

However, after checking the original paper (http://link.springer.com/article/10.1134/S1022795414120138) (I have access to the full paper) I do not see any mention of Z2105, Z2103 or even L23. The data you mentioned is just for R1b-M269. I know for a fact that the Bashkirs have a few percent's of Celtic R1b (U152, L2) and Central Asian R1b-M73. So it's not all Z2103.

Robert6
13-02-15, 11:25
Thanks, I hadn't seen this study published only a few weeks ago. Very timely.

However, after checking the original paper (http://link.springer.com/article/10.1134/S1022795414120138) (I have access to the full paper) I do not see any mention of Z2105, Z2103 or even L23. The data you mentioned is just for R1b-M269. I know for a fact that the Bashkirs have a few percent's of Celtic R1b (U152, L2) and Central Asian R1b-M73. So it's not all Z2103.
From Trofimova work
http://s011.radikal.ru/i318/1502/37/e3e67e8052d7.jpg
http://s011.radikal.ru/i318/1502/37/e3e67e8052d7.jpg

Maciamo
13-02-15, 11:28
From Trofimova work
http://s011.radikal.ru/i318/1502/37/e3e67e8052d7.jpg
http://s011.radikal.ru/i318/1502/37/e3e67e8052d7.jpg

Thanks again. This was not included in the paper in English. Some percentages don't match the paper in English. For example the Mari have 0% of R1b, but have apparently 2% on the graph. Do you have the link to the paper in Russian ?

Here is the translation for those who don't read Cyrillic.

- Bashkirs : 36.2% of Z2105, about 5.2% of M73 and 1.7% of L51/M412
- Chuvash : 2.3% of Z2105
- Komi : 8% of Z2105, 4% of L23* and 3% of L51
- Mari : 2% of L51
- Mordvinians : 7% of Z2103 and 3% of M405/U106/S21
- Kazan Tatars : 1.9% of M73
- Tuymazinsky Tatars : 15% of M405/U106/S21 and 2% of L51
- Udmurts : 21.2% of Z2105
- Besermyans : 3.8% of Z2105

Robert6
13-02-15, 11:40
Thanks again. This was not included in the paper in English.
It is in Russian
http://ibg.anrb.ru/disovet/zashita/2015/02Trofimova/2015_02_TrofimovaDiser.pdf

Robert6
13-02-15, 11:43
The interesting thing is
E1b1b1 G2a J2a J2b in Uralic and Turkic populations of Volga region

Aberdeen
13-02-15, 16:28
Thanks for that info. So, it's as high in parts of Russia as in west Asia. Now the R1b-Z2103 in Samara Yamnas makes more sense.

As of far 3/3 Corded ware have R1a1-M417, 3/3 Bell beaker have R1b(One was tested for P312 and was positive), all but a few of the dozens of Y DNA samples from Bronze-Iron age IE north Asians have R1a, 2/2 Mesolithic Russians have R1b1* and R1a1*, and Upper Palaeolithic MA-1 has R*.

That's an incredible continuum and dominance of Y DNA R in pre-historic ANE-heavy pops of Europe and north Asia. None of the Yamna R1bs were related, all their paternal lineages were not connected recently. This wasn't the result of a recent founder effect, Yamna legitally had a very high amount of R1b.

Who's to say R1b-L11 in west Europe is not descended of the steppe? It's pretty obvious now that we know Yamna and modern eastern Russians have a high amount of L51's brother Z2013.

Actually, there are four Corded Ware samples so far. And while the two German samples from 2600 BC are R1a1, the two Polish samples from 2800 BC are G? and J or I?, which requires some explaining if Corded Ware was a totaling intrusive and possibly complete replacement population as suggested in this new paper.

It is interesting that the most recently discovered German Bell Beaker R1b from 2299-2206 BC is P312S/S116.

Maciamo
13-02-15, 17:40
It is in Russian
http://ibg.anrb.ru/disovet/zashita/2015/02Trofimova/2015_02_TrofimovaDiser.pdf

Thanks again.

It's very interesting to compare the R1a and R1b subclades found among Volga-Uralic peoples.

The Bashkirs are the only ethnic group that lacks western R1a subclades (M458, CTS1211) apart from 1.7% of Z282*. Their dominant R lineages are R1a-Z2125 (31%), R1b-Z2103 (36.2%) and R1b-M73 (5.2%). This is extremely interesting because the Bashkirs occupy the land of the easternmost reaches of the Yamna culture, next to Samara and Orenburg where the Z2103 samples were tested by Haak et al. This would appear to confirm my suspicion that these three haplogroups converged in the Poltavka and/or Sintashta culture before expanding to Central Asia.

Apart from 34.4% of R1a and 43.1% of R1b, the Bashkirs only have N1c (19%) and a little bit of J2a (3.4%).

It's odd that the Balto-Slavic R1a-CTS1211 is found in all ethnic groups in the region except the Bashkirs, as if they had been immune to intermingling with the Slavs (Russians).

The Central European R1a-M458, probably linked to the Corded Ware expansion, is found at low frequencies (1.7 to 7.7%) in all Volga-Ural ethnicities except the Bashkirs, Mari and Tuymazinsky Tatars. This is also reflected in the absence of I1 or I2 among the Bashkirs. Only the Mari and the Udmurts also lack I1 or I2.

The Komi, the Udmurts and the Besermyans, all Uralic people with over 50% of haplogroup N1c, stand out by their complete absence of R1a-Z93, R1a-Z95 or R1a-Z2125. These three ethnic groups and the Mari, another Uralic people, all lack Near Eastern haplogroups G2a, J2a and J2b. The Komi and the Besermyans do have some E1b1b though.

The only Uralic population in the region that differs from all the others are the Mordovians, who display both Germanic (I1, I2a2a), Slavic (32% of R1a-CTS1211, 1.7% of R1a-Z282 and 1.7% of R1a-M458) and apparently Balkanic (E-M78, G2a3b1, J2a, J2b) ancestry. They only have 10% of N1c, not just less than other Uralic speakers, but even less than Turkic speaking ethnic groups (Bashkirs, Chuvash, Tatars). The combination of Germanic, Slavic and Balkanic lineages suggests that the Mordovians could be descended from a branch of the 4th-century Goths from the Chernyakhov culture (http://en.wikipedia.org/wiki/Chernyakhov_culture) in Romania, Moldova and western Ukraine. This is further corroborated by the very strong similarity in names between Moldova and Mordova. Mordovians could therefore be Uralicized Moldovans.

Interestingly, the Chuvash and Tatars carry almost exactly same Germanic (7 to 11% of I1 + some I2-M223), Slavic (mostly R1a-CTS1211, with some R1a-Z282 and R1a-M458) and Balkanic (E-M78, G2a3, I2a1b, J1, J2a, J2b) package as their Mordovian neighbours. That would signify that they also descend from Carpathian Goths. The main difference is that the Chuvash and Tatars have both Uralic N1c1 and also about 10% of Turkic N1c2 (as opposed to the purely Uralic N1c1 of the Mordovians), which explains why they are Turkic speakers today.

My hypothesis is that one group of Carpathian Goths migrated east across Ukraine and settled west of the Volga, where they mixed with the local Uralic (N1c1 + R1a-Z93) speakers, whose language they adopted. In the 7th century, the Bulgars, Turkic speakers from Central Asia, invaded the Volga region and created the Kingdom of Volga Bulgaria (http://en.wikipedia.org/wiki/Volga_Bulgaria) in what is now Chuvashia and Tatarstan. This explains why only the Chuvash and Tatars mixed with them and became Turkic speakers. The Mordovians, who lived further west, with no connection to the Volga, remained Uralic speakers.

bicicleur
13-02-15, 20:38
excuse me for my ignorance,
but aren't Bashkirs speaking a Turkic language?
is this not an area that has been 'visited' many times by different nomadic horseriding tribes?
on the other hand the area is rich in copper ores, so I suppose even if another tribe came to 'visit', the local smiths stayed in the area
I mean, I would expect the Bashkirs to be a mixture of many different ethnicities

Maciamo
13-02-15, 21:28
excuse me for my ignorance,
but aren't Bashkirs speaking a Turkic language?
is this not an area that has been 'visited' many times by different nomadic horseriding tribes?
on the other hand the area is rich in copper ores, so I suppose even if another tribe came to 'visit', the local smiths stayed in the area
I mean, I would expect the Bashkirs to be a mixture of many different ethnicities

Like most Turkic populations in Central Asia today, the Bashkirs were originally an Iranian people who became Turkicized between the 6th and the 11th century. If you look at pictures of Bashkir people, some look very European, while others look very Mongoloid. I suppose that the 58 samples in this study aren't very representative. Lobov et al. 2009 tested 471 Bashkirs separated in eight districts, and there were very big differences in haplogroup composition. For example the Baymaksky and Perm districts had 81% and 84% of R1b-M269, the Abzielilowsky district had 55% of R1b-M73 and 7% of R1b-M269, while the Steribashevky district had no R1b at all ! Western Orenburg had 17% of Mongolian haplogroup C, while eastern Orenburg had 65% of Uralic N1c1. Huge regional variations. What matters here is that I could find a link between R1a-Z2125 and R1b-Z2103, confirming that both are found in the region of origin of the Indo-Aryans and that they therefore could have spread together during the Bronze Age.

Robert6
13-02-15, 22:49
and the Mordovians, another Uralic people, all lack Near Eastern haplogroups G2a, J2a and J2b.
Wait!
The Erzya+Moksha (Mordva) have
1.7% J-12f2
5.1% J2a
10.2% J2b
3.4% G2a
10.2% E1b1b1a

MOESAN
14-02-15, 00:23
I think we cannot rely entirely upon modern populations to track the past, when speaking of relatively low %s - as said, some populations changed language during History but also some mixings occurred at modern times, the Russians being the pole of "redistribution" I think -
Y-E1b is the most surprising, if not from modern Russian (but...) the other "southern" Y-HGroups could go back to the BMAC-Steppics contact (and even osmosis) of Neolithic or Metal Ages (G2a unsure too? we would wait rather G2b? but steppics had conatcs with Tripolje too, so)

bicicleur
14-02-15, 00:45
Like most Turkic populations in Central Asia today, the Bashkirs were originally an Iranian people who became Turkicized between the 6th and the 11th century. If you look at pictures of Bashkir people, some look very European, while others look very Mongoloid. I suppose that the 58 samples in this study aren't very representative. Lobov et al. 2009 tested 471 Bashkirs separated in eight districts, and there were very big differences in haplogroup composition. For example the Baymaksky and Perm districts had 81% and 84% of R1b-M269, the Abzielilowsky district had 55% of R1b-M73 and 7% of R1b-M269, while the Steribashevky district had no R1b at all ! Western Orenburg had 17% of Mongolian haplogroup C, while eastern Orenburg had 65% of Uralic N1c1. Huge regional variations. What matters here is that I could find a link between R1a-Z2125 and R1b-Z2103, confirming that both are found in the region of origin of the Indo-Aryans and that they therefore could have spread together during the Bronze Age.

I guess it will be hard to find out which clades were in the area 4000 years ago, and which clades arrived later
I remembered they spoke Turkic, but with that much R1a / R1b I'm not surprised they look Indo-European
and even the 19 % N1c, seems to be Uralic N1c1a, not Turkic N1c2
the 81 or 84 % R1b-M269 is that M269* or is it just not checked for subclades ?
if it is M269*, they are a special group , this couldn't be coincidence
and with this part of M73 included maybe they were another early split like Anatolian and Tocharian, with some IE language now extinct

If I recall well from reading David Anthony the Indo-Iranians started to spread from east of the southern Urals (with arsenic copper ores) , and not west (with pure copper ores)
But maybe David Anthony also told where they were prior to that

I must say, I tought myself too that R1b-Z2103 was in the vicinity of some R1a-Z93 tribe when the Indo-Iranic expansion started, but as most Indo-Iranian went east and southeast, most R1b-Z2103 went west, they would be Srubnaya
When I think of it : the Z2103 Yamnaya samples found in Samara were 3300-2700 BC, by then Corded Ware got started , so Z2103 must then allready have been situated between R1a-Z283 ot the west and R1a-Z93 to the east

R1a-Z2125 was Andronovo?

Maciamo
14-02-15, 11:03
Wait!
The Erzya+Moksha (Mordva) have
1.7% J-12f2
5.1% J2a
10.2% J2b
3.4% G2a
10.2% E1b1b1a

Sorry I meant the Mari ! I have just added a paragraph on the origins of the Mordovians (who are probably descended from the Goths from Moldova).

Maciamo
14-02-15, 11:28
the 81 or 84 % R1b-M269 is that M269* or is it just not checked for subclades ?
if it is M269*, they are a special group , this couldn't be coincidence
and with this part of M73 included maybe they were another early split like Anatolian and Tocharian, with some IE language now extinct

They didn't test downstream of M269. I originally proposed in 2009 that the Tocharians descended from the M73 branch (although not all M73 are Tocharian, of course). I now think they could have been a mixture of R1a-Z93, R1b-Z2103 and R1b-M73.


R1a-Z2125 was Andronovo?

Andronovo was predominantly R1a-Z93, and most if not all Z93 subclades would have been part and would have appeared during Andronovo period, including Z2125 (Z93>Z94>Z2124>Z2125).

Tomenable
18-02-15, 01:05
I now think they could have been a mixture of R1a-Z93, R1b-Z2103 and R1b-M73.

According to prof. Hui Zhou - though this result is not yet officially published - Tocharian R1a was not Z93:

LINK: http://www.biomedcentral.com/1741-7007/8/15/comments#2168698


Hui Zhou (2014-07-18 16:14) Jilin University

Archaeological and anthropological investigations have helped to formulate two main theories to account for the origin of the populations in the Tarim Basin. The first, so-called “steppe hypothesis”, maintains that the earliest settlers may have been nomadic herders of the Afanasievo culture (ca. 3300-2000 B.C.), a primarily pastoralist culture distributed in the Eastern Kazakhstan, Altai, and Minusinsk regions of the steppe north of the Tarim Basin. The second model, known as the “Bactrian oasis hypothesis”, it maintains that the first settlers were farmers of the Oxus civilization (ca. 2200-1500 B.C.) west of Xinjiang in Uzbekistan, Afghanistan, and Turkmenistan. These contrasting models can be tested using DNA recovered from archaeological bones. Xiaohe cemetery contains the oldest and best-preserved mummies so far discovered in the Tarim Basin, possible those of the earliest people to settle the region. Genetic analysis of these mummies can provide data to elucidate the affinities of the earliest inhabitants.

Our results show that Xiaohe settlers carried Hg R1a1 in paternal lineages, and Hgs H, K, C4, M*in maternal lineages. Though Hg R1a1a is found at highest frequency in both Europe and South Asia, Xiaohe R1a1a more likely originate from Europe because of it not belonging to R1a1a-Z93 branch (our recently unpublished data) which is mainly found in Asians. mtDNA Hgs H, K, C4 primarily distributed in northern Eurasians. Though H, K, C4 also presence in modern south Asian, they immigrated into South Asian recently from nearby populations, such as Near East , East Asia and Central Asia, and the frequency is obviously lower than that of northern Eurasian. Furthermore, all of the shared sequences of the Xiaohe haplotypes H and C4 were distributed in northern Eurasians. Haplotype 223-304 in Xiaohe people was shared by Indian. However, these sequences were attributed to HgM25 in India, and in our study it was not HgM25 by scanning the mtDNA code region. Therefore, our DNA results didn't supported Clyde Winters’s opinion but supported the “steppe hypothesis”. Moreover, the culture of Xiaohe is similar with the Afanasievo culture. Afanasievo culture was mainly distributed in the Eastern Kazakhstan, Altai, and Minusinsk regions, and didn’t spread into India. This further maintains the “steppe hypothesis”.

In addition, our data was misunderstand by Clyde Winters. Firstly, the human remains of the Xiaohe site have no relation with the Loulan mummy. The Xiaohe site and Loulan site are two different archaeological sites with 175km distances. Xiaohe site, radiocarbon dated ranging from 4000 to 3500 years before present, was a Bronze Age site, and Loulan site, dated to about 2000 years before present. Secondly, Hgs H and K are the mtDNA haplogroups not the Y chromosome haplogroups in our study. Thirdly, the origin of Xiaohe people in here means tracing the most recently common ancestor, and Africans were remote ancestor of modern people.
Hui Zhou suggest that the origin of the Xiaohe mummies are of Indo-European origin. In the main article the authors claim that Xiaohe “is different from any other archaeological site of the same period anywhere in the world”. Yet now Hui says the Xiaohe was similar to the Afanaseivo culture. (...) The Afanaseivo culture is characterized by chariots, pottery, timber chamber and rectangular stone enclosure burials, inhumation and creamation. (...)

Tomenable
18-02-15, 02:00
But if not Z93, then what could it be ???

Tomenable
18-02-15, 03:11
BTW - that Clyde Winters, who commented below the paper on Tocharians, is this guy:

https://www.youtube.com/watch?v=DesPF-6-WCE

Arame
09-02-17, 10:41
There are some outdated informations here about Armenians.
The majority of Armenians are L584 (60% of all R1b). Under L584 there is an obvious Armenian cluster with the age 3200 ybp. This basically put an end to speculations that Armenian R1b is related to Hurrians. Because Hurrians became extinct 3200 ago and it was Armenians who were expanding.
It is this lineage. L584 - > PH4150 https://yfull.com/tree/R-Y18781/ formed 4700 ybp, TMRCA 3200 ybp
Notice that formation age 4700 ybp and the expansion age nicely fits into what was known linguistically about Armenians.

The next most frequent is L277.1 (27% of all R1b)
The third is the CTS7763 (~7% of all R1b) which Maciamo thinks is Greek. It was found in LBA Armenia in Kapan town. So it doesn't look Greek or alternatively Greeks entered from Anatolia.
The fourth is PF7562 (~5% of all R1b) This branch look like it is a Hittite. It branches before L23. Upstream case is found in Laz NE Turkey. So it could mean that Hittites came from Maykop?
And then we have Khndzoresk young cluster of L51 and few CTS7822 who recently specificaly tested by admins of Armenian DNA project to see how much impact is there from Balkans. Well not much. 6 cases from 1500 people. So this could be Thracians , Phrygians and others.

Basically this confirms the idea that Armenians entered South Caucasus after the Kura-Araxes ended, bringing Kurganic culture into South Caucasus.
Autosomally they look that they are coming from NW of Black Sea.

Initially their territory was small, but at 1200 BC they profited from the chaotic situation in Near East end expanded their territory. This expansion is visible under the Armenian R1b-L584.

https://s30.postimg.org/eq7b0ytyl/Armenian_R1b_dendrogram_eng.jpg (https://postimg.org/image/eq7b0ytyl/)

IronSide
25-02-17, 23:45
There are some outdated informations here about Armenians.
The majority of Armenians are L584 (60% of all R1b). Under L584 there is an obvious Armenian cluster with the age 3200 ybp. This basically put an end to speculations that Armenian R1b is related to Hurrians. Because Hurrians became extinct 3200 ago and it was Armenians who were expanding.
It is this lineage. L584 - > PH4150 https://yfull.com/tree/R-Y18781/ formed 4700 ybp, TMRCA 3200 ybp
Notice that formation age 4700 ybp and the expansion age nicely fits into what was known linguistically about Armenians.

The next most frequent is L277.1 (27% of all R1b)
The third is the CTS7763 (~7% of all R1b) which Maciamo thinks is Greek. It was found in LBA Armenia in Kapan town. So it doesn't look Greek or alternatively Greeks entered from Anatolia.
The fourth is PF7562 (~5% of all R1b) This branch look like it is a Hittite. It branches before L23. Upstream case is found in Laz NE Turkey. So it could mean that Hittites came from Maykop?
And then we have Khndzoresk young cluster of L51 and few CTS7822 who recently specificaly tested by admins of Armenian DNA project to see how much impact is there from Balkans. Well not much. 6 cases from 1500 people. So this could be Thracians , Phrygians and others.

Basically this confirms the idea that Armenians entered South Caucasus after the Kura-Araxes ended, bringing Kurganic culture into South Caucasus.
Autosomally they look that they are coming from NW of Black Sea.

Initially their territory was small, but at 1200 BC they profited from the chaotic situation in Near East end expanded their territory. This expansion is visible under the Armenian R1b-L584.

https://s30.postimg.org/eq7b0ytyl/Armenian_R1b_dendrogram_eng.jpg (https://postimg.org/image/eq7b0ytyl/)
Very interesting, Proto-Armenians do seem to be mainly R1b-L584 --> PH4150, what about the Persian or Kurdish R1b ? are they L584 ? the Iranian members in the basal subclades project have Armenian surnames, so I haven't seen any true Persian R1b yet, Maciamo in his article on R1b claimes (I don't know if he changed his view) that R1b-L584 correlates more with the Iranian branch of of Indo-Iranian, while L277 is more Indo Aryan, I see a problem in this view because of the large Armenian subclade, if L584 is Iranian then why is the Armenian language an Independent Indo-European branch separate from Iranian ?

this article discusses the origin of the Proto-Armenians, if L584 came from the north of the black sea then which hypothesis is more plausible from the 3 or 4 possibilities the author discusses here:
http://armscoop.com/wp-content/uploads/2011/07/The-Problem-of-Identification-of-the-Proto-Armenians-A-Critical-Review1.pdf

There is an interesting Jewish subclade of L584 --> PF7580 --> FGC14598, which also includes one Lebanese member, could it be the Philistines ? the sea peoples that gave their names to Palestine, if that is the case then we should find modern Palestinians that are positive for this snp, if they ever tested.

Albanian R1b is mostly CTS7822, I don't know what to make of this branch, maybe the original lineage of the Thracians ? it exists in Bulgaria and Romania, the homeland of the Thracians, but also in Russia, Italy, and Spain and so it could have spread with the Goths after they assimilated the native Thracians or Dacians, if we were to assume that then we should expect the Thracian and Dacian languages to be similar to Albanian.
https://en.wikipedia.org/wiki/Classification_of_Thracian#Albanian
https://en.wikipedia.org/wiki/Origin_of_the_Albanians#Thracian_or_Dacian_origin

Finally I agree that PF7562 is better classified as Hittite, it is an old subclade descending directly from R1b-M269, one problem though, its frequency is about 3% in Turkey, isnt that too low to make it the principal lineage of the Hittites ?

Silesian
26-02-17, 01:19
Ossetian is also CTS 7822. Alans and Sarmatians.

ngc598
26-02-17, 12:50
Albanian R1b is mostly CTS7822, I don't know what to make of this branch, maybe the original lineage of the Thracians ? it exists in Bulgaria and Romania, the homeland of the Thracians, but also in Russia, Italy, and Spain and so it could have spread with the Goths after they assimilated the native Thracians or Dacians, if we were to assume that then we should expect the Thracian and Dacian languages to be similar to Albanian.
Language and people are two different things. If you keep them too close together in your mind, things become confused. So far we don't even know if we should place Thracians more to the predominantly R1a people or more to the R1b ones. The lexica of the language itself is more closely related to the baltic languages than with the Albanian and the latter can hardly be closely related to the baltic languages. So we have either a balto-thracian group and an Albanian (Illyrian?) or all three are distinct, (and distinct to the slavic languages as well).

Another point is the time frame. We don't know how old the thracian language branch really is, therefore we can't place Thracians in a specific Haplogroup. You can see how tricky this is with the slavic languages. They converge roughly 1500-2000 years ago in the Balkans, but the 'west slavic' languages are spoken by dominantly M458 people, who split from the Balkan people at least 2000 years before the 'origin' of the slavic languages. The consequences of this is that you can't call Poles or Sorbs Slavs even if they speak a slavic language. The same goes for Romanians, you can't call them Romans as well, just because they speak a Romance language. So language and people often are separated in space and time and assigning haplogroups to languages is full of errors.

Arame
27-02-17, 10:37
IronSide

Well I can't answer instead of Maciamo. He has a theory that Colchian culture in South Caucasus is Armenian. But I don't know any scholar who proposed such a strange idea. Colchian culture was a Kartvelian culture with Aryan/Scythian superstrate/elite. Most probably the reason why Ossetians have high level of G2 is due to this long period of cohabitation.

One thing I can say judging with the Big Ys we are getting it is very unlikely that L584 and L277 comes from Central Asia. Most probably they entered via Caucasus after 2400BC.

Here are new samples of Armenians under L277
https://yfull.com/tree/R-Y4364/

Notice that 3 Armenians fall under a 2600 year old branch. The other one is in more upstream position.
Most probably the majority of Armenians L277 will fall under this 2600 year old branch. The age is not definitive. This L277 branch is of course not Armenian ( except that one branch ) but linking it with Iranian expansions is also problematic. The reason is that the place from where it expands is in NW Iran/S Caucasus, and it starts to expand _before_ the arrival of Iranians.
So who could be related to this branch?

My theory is that this is a branch of now extinct IE language. This was the Gutians. Very little is known about Gutian language but some scholars hypothesized that they were IE group with a language similar to Tocharian ( one of king names Tirigan has a tocharian like word -gan at the end, making the etymology of this name like Tir Given similar to Iranian Tiridat ) The presence of L277 in Mesopotamia and Levant also favours this theory. The name of Gutians is similar to Goths, Getae. This name are derived from IE *wet - > get meaning river people.

Also Kurds consider Gutians as their ancestors, and it seems that the most frequent R1b among Kurds is the L277 one.

As for Armenian ethnogenesis the best theory that fits into genetic data is the "Dragon Stone" theory proposed recently by Hrach Martirosyan et al. This theory is already partially proven by genetic data. We are just waiting to see Trialeti culture kurgan burials. If they were L584 then a lot off things will be solved. I will present this theory and my view on L584 tomorrow.

Arame
01-03-17, 10:34
IronSide

I couldn't get that paper on Dragon Stones in English. It's not yet published.
So I will tell my opinion based on genetic data.
In essence it is a mix of Etiuni hypothesis and Hayassa one.A migration from modern Armenia Republic (Etiuni) to Upper parts of Euphrates and formation there of Hayassa, which was a multiethnic state with Hattic like substrate. Later when Hayassa disappears then the Mushki and Urumu ( correct reading is probably Aramu ) expands from ex Hayassa Upper Euphrate to South and South West. In South they are defeated by Assyria. And they continue to move to West where later we see Greek Moschois there. This Mushkis expansions has a strong archaeological basis in the form of potteries.
All this events are mostly related to L584. So L584 is not only Armenian but also Mushki/Moschoi/Meshekh. Also most probably L584 played a role in the formation of Manneans. We have aDNA from NW Iran in Mannean lands. 900 BC. Btw that aDNA doesn't show any hint that it came from Central Asia, Andronovo / BMAC like Maciamo suggests.

Concerning L584 homeland after Steppe. My theory ( we will check that with aDNA in coming years) is that L584 appeared in Trialeti culture. Then it expanded West. Formed Hayassa. Hayassa collapses during LBA crisis and we have Mushki and Urumu expansion that we can easily link to that 3200 year old cluster under L584. One branch moved quite early in South. Lake Urmia, North Levant. Jews probably get their L584 either from Manneans, or Mushki/Meshekhs ( attested in Bible as a son of Aram ) Or alternatively from Philistineans.

Btw I didn't answer Your question about Iranian R1b. I expect them to be very diverse. L584 in theory shouldn't be a majority in Iran, but the L277. This L277 probably expanded with Saka and Massagetes later when Gutians where Iranised. Also there should be a hotspot off Z2110 in Zagros in correlation with E-V13. Various branches of 2106 are expected. And maybe some P312. Here is a link on Kurdish Y dna.

http://corduene.blogspot.am/2017/01/december-2016-update.html

AlGreen
25-10-17, 16:32
Sorry I meant the Mari ! I have just added a paragraph on the origins of the Mordovians (who are probably descended from the Goths from Moldova).

30%+ Balkans + I2a and R1b-Z2103 could safely assume an indigenous Lower Danube-Carpathian population of Dacian-Getae + Celto-Germanic Bastarne and/or Goths + Sarmat / Slavic, depending on age.

If we talk about Chernyakov - Santana culture, then probably the reason of Mordvins displacement could be due to Huns;
Possibly taken into slavery along with a local mix Celto-Germanic-Sarmat. found at NW of Black Sea during that time.

There is also the hypothesis that the Mordovians could be direct descendants from Tyragetae, Tysagetae - not far from their attested spot or Samo-Getae (given also the R1a and I1). Looks like Mordovian folklore is also similar to Moldovans.

Mehmet
13-03-18, 06:23
hi. I am Chinese who have just tested Y Chromosome and get the result in YFULL which is R1b-Z2106-CTS8966 with Sample ID:ELT50011

IronSide
13-03-18, 15:19
hi. I am Chinese who have just tested Y Chromosome and get the result in YFULL which is R1b-Z2106-CTS8966 with Sample ID:ELT50011

Cool. Which region of china are you from if I may ask ?

Mehmet
14-03-18, 04:37
Cool. Which region of china are you from if I may ask ?

I am now living in eastern China.

IronSide
14-03-18, 10:28
I am now living in eastern China.

Sorry I didn't phrase my question correctly. Where were you born in China ? Which region .. I'm guessing north-west China ?

Mehmet
15-03-18, 17:45
I borned in eastern China.

IronSide
15-03-18, 19:04
I borned in eastern China.

Sorry if this is personal, but are you Muslim ? or of Muslim background ?

Mehmet
16-03-18, 18:09
actually I am Hui people in China who is minor ethnic group primarily migrated from Central or West Asia due to Mongol's invasion.

atalay34
10-10-18, 01:56
Can anybody explain basically when was z2301 born and where. What is the place of origin of z2103?

Saetrus
10-10-18, 02:37
Can anybody explain basically when was z2301 born and where. What is the place of origin of z2103?

The oldest R1b-Z2103 we have is from Hajji Firuz in Iran ca. 5650 BC, anything other than that is speculative.

ToBeOrNotToBe
10-10-18, 05:28
The oldest R1b-Z2103 we have is from Hajji Firuz in Iran ca. 5650 BC, anything other than that is speculative.

I forgot about that guy - that's bloody old, are we sure the C14 is correct?

Pip
13-10-18, 14:08
It is indeed old, considering yfull estimates that Z2103 began forming in 4,100 BC and is 95% confident that it is no older than 4,800 BC. Mistakes in classification are not unknown, although my estimates are that formational Z2103 could well be as old as 5,500 BC if some curiously divergent outliers are incorporated into the analysis.

There are ten Z2103-equivalent SNPs that could have come together over a thousand or more years, so what is considered to be the 'birthplace' of Z2103 is probably a very wide area, especially if its long sequence of bearers were nomadic.

The relevant data is limited, but suggests that the bearers of formational Z2103 most likely existed predominantly over a wide arc running from the Caucasus region to Poland. Its two surviving branches appear to have separated fairly early on - (i) a Western branch that spread around Germany and the Balkans before withering and being subsumed into other populations (e.g. R1a-dominated Corded Ware), and (ii) an Eastern branch that retreated into relative isolation around the Caucasus and Caspian before merging later into mainly Caucasian populations. (There would also have been other branches that died out - one isolated early Northern Spanish sample, for instance, has a curious autosomal similarity to Balkan Z2103.)

I am interested in the modern distribution of Z2103 South of the Caucasus, and the degree to which it might descend from Eastern Yamnaya, Maykop and/or Hittite populations, if anyone has further information on this?

Pip
13-10-18, 17:40
I have just checked on the Hajji Firuz R1b-Z2103 sample dated 5,650 BC.

The same study shows a sample of R1b-U106 from Iron Gates Serbia dated 8,885 BC. However, the subclade of U106 identified for this sample (R1b1a1a2a1a1b1a1a) is estimated by yfull to have a formation date of only 700 BC.

My understanding is that the readings in this study are only estimated, based on calls. As with many reports in relation to ancient samples, the reliability of this reading is perhaps questionable.

Silesian
15-10-18, 01:03
It is indeed old, considering yfull estimates that Z2103 began forming in 4,100 BC and is 95% confident that it is no older than 4,800 BC. Mistakes in classification are not unknown, although my estimates are that formational Z2103 could well be as old as 5,500 BC if some curiously divergent outliers are incorporated into the analysis.

There are ten Z2103-equivalent SNPs that could have come together over a thousand or more years, so what is considered to be the 'birthplace' of Z2103 is probably a very wide area, especially if its long sequence of bearers were nomadic.

The relevant data is limited, but suggests that the bearers of formational Z2103 most likely existed predominantly over a wide arc running from the Caucasus region to Poland. Its two surviving branches appear to have separated fairly early on - (i) a Western branch that spread around Germany and the Balkans before withering and being subsumed into other populations (e.g. R1a-dominated Corded Ware), and (ii) an Eastern branch that retreated into relative isolation around the Caucasus and Caspian before merging later into mainly Caucasian populations. (There would also have been other branches that died out - one isolated early Northern Spanish sample, for instance, has a curious autosomal similarity to Balkan Z2103.)

I am interested in the modern distribution of Z2103 South of the Caucasus, and the degree to which it might descend from Eastern Yamnaya, Maykop and/or Hittite populations, if anyone has further information on this?

An online active ydna/culture map,with ancient R1b-Z2103>Z2108+Z2109+Z2110+ samples [Yamnaya,Afansievo,Polatavka,Catacombe,Vucedol,Bell Beaker,Sintashta?,Sarmatian,Scythian etc...]

http://images.devs-on.net/Image/67vhlt2ViGcKn2jK-Region.png

http://images.devs-on.net/Image/67vhlt2ViGcKn2jK-Region.png

Pip
17-10-18, 00:21
Eastern R1b-Z2103 appears to have developed in the Caucasus region at around the same time as the Dolmen culture arose in this region. Might the heavy concentration of modern Z2103 around the Western Caucasus be linked to this, rather than preceding Steppe cultures? If not R1b-Z2103, which other y-DNA would the Dolmen people have borne?

Pip
20-10-18, 13:31
The modern distribution of R1b-Z2103 exhibits its greatest diversity in two areas - the Southern Caucasus and East Central Europe. SNPs on yfull's database suggest a most likely formation zone for Z2103 in the Caucasus and STRs/SNPs on FTDNA's database suggest a most likely formation zone in East Central Europe, but there is insufficient data to arrive at a firm conclusion one way or the other.

There is little evidence in modern distributions to indicate a significant early development of surviving Z2103 populations anywhere across the vast area of the Steppe - it seems that either the early Z2103 Steppe populations largely evacuated the area or were otherwise eradicated from it, or that the Steppe was principally a migratory route for them between the Caucasus and East Central Europe.

Autosomally, early Z2103 samples (3,000-2,800 BC) across Eastern Europe (Balkans, Ukraine and Russia) are strikingly diverse, suggesting that yfull's TMRCA of 3,600 BC for Z2103 might be an underestimate. A combined analysis of SNPs and STRs would suggest a TMRCA of something more like 4,400 BC.

However, the first significant developmental stages of Z2103's Western and Eastern branches each appear to date to only approximately 3,200 BC. Eastern Z2103 is curious; both yfull's and FTDNA's databases seem to indicate the same thing - that its first significant developments were in Armenia, even though samples from this area at that time show little sign of the autosomal DNA that is core to early Z2103 samples. My suggestions are that (i) the ancestors of today's eastern Z2103 were located just to the West of where it now has its most diverse presence - along the low-lying coastal area from Circassia to North Eastern Turkey, and (ii) Z2103 spread into Armenia from the West just as R1a-Z93 moved into it from the North East during the early third millennium BC.

halfalp
20-10-18, 15:09
The modern distribution of R1b-Z2103 exhibits its greatest diversity in two areas - the Southern Caucasus and East Central Europe. SNPs on yfull's database suggest a most likely formation zone for Z2103 in the Caucasus and STRs/SNPs on FTDNA's database suggest a most likely formation zone in East Central Europe, but there is insufficient data to arrive at a firm conclusion one way or the other.

There is little evidence in modern distributions to indicate a significant early development of surviving Z2103 populations anywhere across the vast area of the Steppe - it seems that either the early Z2103 Steppe populations largely evacuated the area or were otherwise eradicated from it, or that the Steppe was principally a migratory route for them between the Caucasus and East Central Europe.

Autosomally, early Z2103 samples (3,000-2,800 BC) across Eastern Europe (Balkans, Ukraine and Russia) are strikingly diverse, suggesting that yfull's TMRCA of 3,600 BC for Z2103 might be an underestimate. A combined analysis of SNPs and STRs would suggest a TMRCA of something more like 4,400 BC.

However, the first significant developmental stages of Z2103's Western and Eastern branches each appear to date to only approximately 3,200 BC. Eastern Z2103 is curious; both yfull's and FTDNA's databases seem to indicate the same thing - that its first significant developments were in Armenia, even though samples from this area at that time show little sign of the autosomal DNA that is core to early Z2103 samples. My suggestions are that (i) the ancestors of today's eastern Z2103 were located just to the West of where it now has its most diverse presence - along the low-lying coastal area from Circassia to North Eastern Turkey, and (ii) Z2103 spread into Armenia from the West just as R1a-Z93 moved into it from the North East during the early third millennium BC.

Using modern population to explain ancient migration is the wrongest thing ever. Because like showing in India, their endogamism have created such founder effect that even if R1a-Z93 is found in prehistoric context in Eastern Europe / Central Asia, modern India is such a totally other population. Underhill did the same mistake by infering R1a must expanded from Iran because diversity was higher there. All those datas are absolutely missleading the reality of prehistoric migrations. The Pontic Zone meaning the Black Sea is an Horseshoe with the Caucasus at his Base. Meaning that, looking at ancient datas from Pontic Steppe were R1b and R1a originate at one point, their modern distribution and modern diversity gonna for sure being around Eastern Europe ( Russia, Ukraine ) -> Romania -> Bulgaria -> Turkey -> South Caucasus -> Iran. We still gonna have to listen to their everlasting hypothesis, until ancient samples only gonna confirm the north eurasian origin and expansion of R-related lineage. We should start to ask to some people especially Daamgard, why he dont test for y-dna some very important samples in relation with thoses expansions.

Silesian
20-10-18, 15:39
Using modern population to explain ancient migration is the wrongest thing ever. Because like showing in India, their endogamism have created such founder effect that even if R1a-Z93 is found in prehistoric context in Eastern Europe / Central Asia, modern India is such a totally other population. Underhill did the same mistake by infering R1a must expanded from Iran because diversity was higher there. All those datas are absolutely missleading the reality of prehistoric migrations. The Pontic Zone meaning the Black Sea is an Horseshoe with the Caucasus at his Base. Meaning that, looking at ancient datas from Pontic Steppe were R1b and R1a originate at one point, their modern distribution and modern diversity gonna for sure being around Eastern Europe ( Russia, Ukraine ) -> Romania -> Bulgaria -> Turkey -> South Caucasus -> Iran. We still gonna have to listen to their everlasting hypothesis, until ancient samples only gonna confirm the north eurasian origin and expansion of R-related lineage. We should start to ask to some people especially Daamgard, why he dont test for y-dna some very important samples in relation with thoses expansions.
One of the most successful military armies-Mongols



War
Death Range
Location


Mongol conquests
40,000,000–70,000,000
Eurasia


Taiping Rebellion
20,000,000–100,000,000
China


Three Kingdoms War
36,000,000–40,000,000
China


Conquest of the Americas
8,400,000–137,750,000
Americas





The Battle of Samara Bend (Russian (https://en.wikipedia.org/wiki/Russian_language): Монгольско-булгарское сражение, lit. (https://en.wikipedia.org/wiki/Literal_translation) 'Mongolian-Bulgarian battle') or the Battle of Kernek was the first battle between Volga Bulgaria (https://en.wikipedia.org/wiki/Volga_Bulgaria) and the Mongols (https://en.wikipedia.org/wiki/Mongol_Empire), probably one of the first skirmishes or battles the Mongols lost. It took place in autumn 1223, at the southern border of Volga Bulgaria. The Bulgars retreated and the Mongols pursued them. Then the main Bulgar forces ambushed the Mongols.

Samara- region has been the home of R1b-Z2103 [Yamnaya/Poltovka/Catacombe to south,Sintashta/ Sarmatians, and Bashkirs]for a minimum of 5500YBP+/- if not older.

Any trace of Ghenis khan descendants in the region?


DNA evidence[edit (https://en.wikipedia.org/w/index.php?title=Descent_from_Genghis_Khan&action=edit&section=5)]

Numerous studies by teams of biochemists, based on the Y-DNA of modern descendants of Genghis Khan, have indicated that Genghis Khan may have belonged to a subclade of Haplogroup C-M217 (https://en.wikipedia.org/wiki/Haplogroup_C-M217) (C2) such as C-F4002 (C2b1a3).[8] (https://en.wikipedia.org/wiki/Descent_from_Genghis_Khan#cite_note-8)
Proponents of the theory regarding C-M217 have put forward hypothetical Y-DNA profiles, such as a 25 Marker "Genghis Khan" profile released by Family Tree DNA (https://en.wikipedia.org/wiki/Family_Tree_DNA):

https://en.wikipedia.org/wiki/Descent_from_Genghis_Khan










http://images.devs-on.net/Image/3P8zNAAf02AGiYkq-Region.png]

halfalp
20-10-18, 15:55
Not sure to get your message.

Pip
20-10-18, 16:42
The thread is about R1b-Z2103 (rather than R1a-Z93 or R-related lineages in general), and about its modern (rather than its ancient) distribution. However, regarding the points made, I have no opinion to express on whether R-related lineages had a North Eurasian origin. And there is no reason to suppose that various branches of R1a did not expand both inside and far outside of Iran over their very long periods of formation and development; to try to locate a single pinpoint is futile.

Data from modern populations provide a lot of detailed and accurate information that could relate to the most likely migrations of surviving branches. On the other hand, data from ancient samples is relatively sparse, commonly relates to lineages that have died out, and is frequently beset by classification errors. However, no information should be wholly disregarded; everything should be brought into consideration.

Regarding modern populations of Z2103 specifically, I do not see anything in the data to demonstrate a North Eurasian aspect to the development of its surviving branches. There is a cluster of related early branches that coalesces around Armenia and a similar cluster that coalesces West of the Dniester - these are the branches that survive and thrive today.

If anyone has any information to suggest that the immediate common ancestors of the Eastern branch of modern Z2103 most likely lived in an area other than the south eastern shores of the Black Sea around 3,000 BC, then I would be interested to hear about it.

Furthermore, given that (i) Yamnayan Z2103 samples exhibit a highly uniform and stable autosomal mix, and (ii) non-Yamnayan European Z2103 samples are very diverse autosomally (both from Yamnaya and from each other), is there any more likely explanation for this than that their DNA developed separately within different European populations at an earlier date, rather than as part of a single Yamnayan incusion at a later date?

halfalp
20-10-18, 16:46
The thread is about R1b-Z2103 (rather than R1a-Z93 or R-related lineages in general), and about its modern (rather than its ancient) distribution. However, regarding the points made, I have no opinion to express on whether R-related lineages had a North Eurasian origin. And there is no reason to suppose that various branches of R1a did not expand both inside and far outside of Iran over their very long periods of formation and development; to try to locate a single pinpoint is futile.

Data from modern populations provide a lot of detailed and accurate information that could relate to the most likely migrations of surviving branches. On the other hand, data from ancient samples is relatively sparse, commonly relates to lineages that have died out, and is frequently beset by classification errors. However, no information should be wholly disregarded; everything should be brought into consideration.

Regarding modern populations of Z2103 specifically, I do not see anything in the data to demonstrate a North Eurasian aspect to the development of its surviving branches. There is a cluster of related early branches that coalesces around Armenia and a similar cluster that coalesces East of the Dniester - these are the branches that survive and thrive today.

If anyone has any information to suggest that the immediate common ancestors of the Eastern branch of modern Z2103 most likely lived in an area other than the south eastern shores of the Black Sea around 3,000 BC, then I would be interested to hear about it.

Furthermore, given that (i) Yamnayan Z2103 samples exhibit a highly uniform and stable autosomal mix, and (ii) non-Yamnayan European Z2103 samples are very diverse autosomally (both from Yamnaya and from each other), is there any more likely explanation for this than that their DNA developed separately within different European populations at an earlier date, rather than as part of a single Yamnayan incusion at a later date?

Yes there is absolutely a reason... because it's wrong. No paleolithic or neolithic to date sample fro all that region were R1. No sign of Basal Eurasian into the ancient R1a sample that we got.

halfalp
20-10-18, 16:51
The thread is about R1b-Z2103 (rather than R1a-Z93 or R-related lineages in general), and about its modern (rather than its ancient) distribution. However, regarding the points made, I have no opinion to express on whether R-related lineages had a North Eurasian origin. And there is no reason to suppose that various branches of R1a did not expand both inside and far outside of Iran over their very long periods of formation and development; to try to locate a single pinpoint is futile.

Data from modern populations provide a lot of detailed and accurate information that could relate to the most likely migrations of surviving branches. On the other hand, data from ancient samples is relatively sparse, commonly relates to lineages that have died out, and is frequently beset by classification errors. However, no information should be wholly disregarded; everything should be brought into consideration.

Regarding modern populations of Z2103 specifically, I do not see anything in the data to demonstrate a North Eurasian aspect to the development of its surviving branches. There is a cluster of related early branches that coalesces around Armenia and a similar cluster that coalesces West of the Dniester - these are the branches that survive and thrive today.

If anyone has any information to suggest that the immediate common ancestors of the Eastern branch of modern Z2103 most likely lived in an area other than the south eastern shores of the Black Sea around 3,000 BC, then I would be interested to hear about it.

Furthermore, given that (i) Yamnayan Z2103 samples exhibit a highly uniform and stable autosomal mix, and (ii) non-Yamnayan European Z2103 samples are very diverse autosomally (both from Yamnaya and from each other), is there any more likely explanation for this than that their DNA developed separately within different European populations at an earlier date, rather than as part of a single Yamnayan incusion at a later date?

Omg... so your argument is that ancient R1b-Z2103 from Pontic Steppe is different autosomally than modern Armenian branch. Bravo. No y-dna R1b-Z2103 was found in important sites from Armenia, neither found in Kura-Araxes or Maikop proper, wich does not make sense, there was no hiding R1b-Z2103 in prehistoric Armenia that resurface in modern times. Wich basically means that modern Armenian R1b-Z2103 have an obvious steppic origin linking with the way more obvious fact that Armenia and Armenian are a surviving branch of IE language and culture.

Silesian
20-10-18, 17:40
Not sure to get your message.


https://www.youtube.com/watch?v=_BCXAUEBgO4

markod
20-10-18, 17:41
I have just checked on the Hajji Firuz R1b-Z2103 sample dated 5,650 BC.

The same study shows a sample of R1b-U106 from Iron Gates Serbia dated 8,885 BC. However, the subclade of U106 identified for this sample (R1b1a1a2a1a1b1a1a) is estimated by yfull to have a formation date of only 700 BC.

My understanding is that the readings in this study are only estimated, based on calls. As with many reports in relation to ancient samples, the reliability of this reading is perhaps questionable.

I think the Z2103 in Hajji Firuz has been confirmed by others. It hasn't been dated directly however.

halfalp
20-10-18, 18:44
https://www.youtube.com/watch?v=_BCXAUEBgO4

I still dont get it. Obviously in an ancient context, R1a-Z93 and IE languages ultimately came from Pontic Steppe. But the modern population of India, her diversity, her endogamism would make it R1a-Z93 more likely to be local than to come from elswhere. Do my point, you cannot use modern dna to explain ancient migrations, you just cant, even if you have obviously more datas than ancient datas.

markod
20-10-18, 19:25
I still dont get it. Obviously in an ancient context, R1a-Z93 and IE languages ultimately came from Pontic Steppe. But the modern population of India, her diversity, her endogamism would make it R1a-Z93 more likely to be local than to come from elswhere. Do my point, you cannot use modern dna to explain ancient migrations, you just cant, even if you have obviously more datas than ancient datas.

Z93 basal diversity always pointed to an origin in Central Asia, as reported in Underhill (2014).

Silesian
20-10-18, 19:34
I still dont get it. Obviously in an ancient context, R1a-Z93 and IE languages ultimately came from Pontic Steppe. But the modern population of India, her diversity, her endogamism would make it R1a-Z93 more likely to be local than to come from elswhere. Do my point, you cannot use modern dna to explain ancient migrations, you just cant, even if you have obviously more datas than ancient datas.
Do you agree with yfull that Steppe R1b-z2103 is older than some Steppe R1a branches?


R-L23 (https://www.yfull.com/tree/R-L23/)PF6404 * L478/PF6403 * L23/S141/PF6534formed 6400 ybp, TMRCA 6100 ybp]info (https://www.yfull.com/branch-info/R-L23/)

R-L23* (https://www.yfull.com/tree/R-L23*/)
R-Z2103 (https://www.yfull.com/tree/R-Z2103/) Y4371/Z8128/M12149 * S20902/Z8130 * CTS9416+7 SNPs formed 6100 ybp, TMRCA 5600 ybpinfo (https://www.yfull.com/branch-info/R-Z2103/)

R-Z2103* (https://www.yfull.com/tree/R-Z2103*/)
R-PF331 (https://www.yfull.com/tree/R-PF331/)FGC35088 * PF331formed 5600 ybp, TMRCA 4700 ybpinfo (https://www.yfull.com/branch-info/R-PF331/)

https://www.yfull.com/static/css/img/person.pngid:YF07907
ITA [IT-PR]



R-Z283 (https://www.yfull.com/tree/R-Z283/)Z662/CTS11197/PF6225 * Z283/S339/PF6217formed 5000 ybp, TMRCA 4900 ybpinfo (https://www.yfull.com/branch-info/R-Z283/)

R-Z645 (https://www.yfull.com/tree/R-Z645/)Z646/CTS6596/M713/S346 * Z645/S224/PF6162/V1754 * Z650/CTS9754/PF6206/M750/V3726]+5 SNPsformed 5500 ybp, TMRCA 5000 ybpinfo (https://www.yfull.com/branch-info/R-Z645/)

R-Z645* (https://www.yfull.com/tree/R-Z645*/)
R-Z93 (https://www.yfull.com/tree/R-Z93/)Z2479/M746/S4582/V3664 * FGC77882 * Z93/F992/S202formed 5000 ybp, TMRCA 4700 ybpinfo (https://www.yfull.com/branch-info/R-Z93/)

https://www.yfull.com/static/css/img/person.pngid:ERS256938
[COLOR=#777777]ITA [IT-CA]

R-Z93* (https://www.yfull.com/tree/R-Z93*/)




https://www.yfull.com/tree

halfalp
20-10-18, 19:39
Z93 basal diversity always pointed to an origin in Central Asia, as reported in Underhill (2014).

Do you have a quote about Z93 basal diversity pointing to Central Asia in the Paper? I only see the following.

Y-STR haplotype networks and diversityWe genotyped a subset of 1355 R1a samples for 10–19 Y-chromosome STR loci (Supplementary Table 3 (https://www.nature.com/articles/ejhg201450#s1)) and constructed networks for both hg R1a-Z282 and hg R1a-Z93 (Supplementary Figure 1 (https://www.nature.com/articles/ejhg201450#s1) and Supplementary Figure 2 (https://www.nature.com/articles/ejhg201450#s1)). Although we could assign haplotypes to various haplogroups, power to identify substructure within hg R1a-M198 was limited, consistent with previous work.22 (https://www.nature.com/articles/ejhg201450#ref22), 52 (https://www.nature.com/articles/ejhg201450#ref52) Although haplotype diversity is generally very high (H>0.95) in all haplogroups (Supplementary Table 3 (https://www.nature.com/articles/ejhg201450#s1)), lower diversities occur in south Siberian paragroup R1a-Z93* (H=0.921), in Jewish R1a-M582 (H=0.844) and in Roma R1a-M780 (H=0.759), consistent with founder effects that are evident in the network patterns for these populations (Supplementary Figure 2 (https://www.nature.com/articles/ejhg201450#s1)).

Wich make them believe that Z-93 origin was likely south asia.

halfalp
20-10-18, 19:41
Diversity of an haplogroup gonna be higher into a geographical area comprizing multiple and diverse lineages isn't? So obviously even I2 could be more diverse in Kurdistan.

halfalp
20-10-18, 19:44
Do you agree with yfull that Steppe R1b-z2103 is older than some Steppe R1a branches?



https://www.yfull.com/tree

What do i know?

markod
20-10-18, 19:45
Do you have a quote about Z93 basal diversity pointing to Central Asia in the Paper? I only see the following.

Y-STR haplotype networks and diversity

We genotyped a subset of 1355 R1a samples for 10–19 Y-chromosome STR loci (Supplementary Table 3 (https://www.nature.com/articles/ejhg201450#s1)) and constructed networks for both hg R1a-Z282 and hg R1a-Z93 (Supplementary Figure 1 (https://www.nature.com/articles/ejhg201450#s1) and Supplementary Figure 2 (https://www.nature.com/articles/ejhg201450#s1)). Although we could assign haplotypes to various haplogroups, power to identify substructure within hg R1a-M198 was limited, consistent with previous work.22 (https://www.nature.com/articles/ejhg201450#ref22), 52 (https://www.nature.com/articles/ejhg201450#ref52) Although haplotype diversity is generally very high (H>0.95) in all haplogroups (Supplementary Table 3 (https://www.nature.com/articles/ejhg201450#s1)), lower diversities occur in south Siberian paragroup R1a-Z93* (H=0.921), in Jewish R1a-M582 (H=0.844) and in Roma R1a-M780 (H=0.759), consistent with founder effects that are evident in the network patterns for these populations (Supplementary Figure 2 (https://www.nature.com/articles/ejhg201450#s1)).

Wich make them believe that Z-93 origin was likely south asia.

They literally call it the Siberian paragroup ;)

See the joining network:

https://media.nature.com/original/nature-assets/ejhg/journal/v23/n1/extref/ejhg201450x2.pdf

markod
20-10-18, 19:47
Diversity of an haplogroup gonna be higher into a geographical area comprizing multiple and diverse lineages isn't? So obviously even I2 could be more diverse in Kurdistan.

Generally that's how it works, but obviously these methods won't take into account how amenable a region is to the preservation of haplotypes. Perhaps the Pontic-Caspian steppe in particular (more so than the Siberian plain) saw massive Y-DNA turnovers several times.

halfalp
20-10-18, 19:47
They literally call it the Siberian paragroup ;)

See the joining network:

https://media.nature.com/original/nature-assets/ejhg/journal/v23/n1/extref/ejhg201450x2.pdf

Yes but their argument is that R1a must expand from Iran because of the highest diversity, while the Siberian paragroup R1a Z-93 have the lowest diversity of the other paragroups. Diversity and Genetic Drift are more likely to happened in very populous and genetically diverse regions like middle-east and india, than in very low density population like central asia.

Pip
20-10-18, 20:10
Yes there is absolutely a reason... because it's wrong. No paleolithic or neolithic to date sample fro all that region were R1. No sign of Basal Eurasian into the ancient R1a sample that we got.
No y-DNA has been found at all in many ancient cultures.
Of course, at some point over the 23,000 years between R1's formation and the end of the Neolithic, at least one R1 man would have wandered down the Caspian coastline from the Steppe to Iran.
R1 crops up in Neolithic Spain and in Equatorial Africa, so why would Iran be out of the question?

Pip
20-10-18, 20:16
Omg... so your argument is that ancient R1b-Z2103 from Pontic Steppe is different autosomally than modern Armenian branch. Bravo. No y-dna R1b-Z2103 was found in important sites from Armenia, neither found in Kura-Araxes or Maikop proper, wich does not make sense, there was no hiding R1b-Z2103 in prehistoric Armenia that resurface in modern times. Wich basically means that modern Armenian R1b-Z2103 have an obvious steppic origin linking with the way more obvious fact that Armenia and Armenian are a surviving branch of IE language and culture.
My question was Furthermore, given that (i) Yamnayan Z2103 samples exhibit a highly uniform and stable autosomal mix, and (ii) non-Yamnayan European Z2103 samples are very diverse autosomally (both from Yamnaya and from each other), is there any more likely explanation for this than that their DNA developed separately within different European populations at an earlier date, rather than as part of a single Yamnayan incusion at a later date?
It is about the diversity in European Z2103, not Armenian Z2103.

halfalp
20-10-18, 20:24
No y-DNA has been found at all in many ancient cultures.
Of course, at some point over the 23,000 years between R1's formation and the end of the Neolithic, at least one R1 man would have wandered down the Caspian coastline from the Steppe to Iran.
R1 crops up in Neolithic Spain and in Equatorial Africa, so why would Iran be out of the question?

Fair enough, and what would be is Elbrouz-like autosomal dna? and the mtdna haplogroups coming with him?

" R1 " crops in paleolithic Italy and in modern " not equatorial " Africa, do you feel a subtility? The same group of modern African R1b pops in mesolithic Balkans and neolithic Ukraine / Russia. No subtility yet?

Pip
20-10-18, 20:25
I think the Z2103 in Hajji Firuz has been confirmed by others. It hasn't been dated directly however.
Even if not confirmed Z2103, its calls would at least suggest formative R1b-M269 to be its most likely haplogroup and in any case to indicate that Steppic DNA would have moved into Iran prior to the Chalcolithic.

halfalp
20-10-18, 20:28
My question was Furthermore, given that (i) Yamnayan Z2103 samples exhibit a highly uniform and stable autosomal mix, and (ii) non-Yamnayan European Z2103 samples are very diverse autosomally (both from Yamnaya and from each other), is there any more likely explanation for this than that their DNA developed separately within different European populations at an earlier date, rather than as part of a single Yamnayan incusion at a later date?
It is about the diversity in European Z2103, not Armenian Z2103.

Yes its called multiple founder effects. R1b-Z2103 lost predominance everywhere after the Bronze Age, while it survive pretty good in Anatolia and Armenia. Thanks to mountaneous regions and large genetic diversity. Europe is a geographical bottleneck, while middle-east have more holes than swiss cheese.

halfalp
20-10-18, 20:33
Even if not confirmed Z2103, its calls would at least suggest formative R1b-M269 to be its most likely haplogroup and in any case to indicate that Steppic DNA would have moved into Iran prior to the Chalcolithic.

Probably all R1b subgroups from V-88 to M269 to Z2103 to L23 journey from Balkans or Eastern Europe to the south at some point. It dont even have to be linked with IE languages or Yamnaya to be a fact.

Pip
20-10-18, 20:53
Yes but their argument is that R1a must expand from Iran because of the highest diversity, while the Siberian paragroup R1a Z-93 have the lowest diversity of the other paragroups. Diversity and Genetic Drift are more likely to happened in very populous and genetically diverse regions like middle-east and india, than in very low density population like central asia.
Low density populations would only have lower average STR diversity to the extent that there would have been periods of time when only one man with the haplogroup lived to reach reproductive age.
As long as there are two men with the haplogroup, their descendants' STRs would mutate independently and the diversity between the STRs for the two groups would continue to increase at the same average rate.

Pip
20-10-18, 21:14
Fair enough, and what would be is Elbrouz-like autosomal dna? and the mtdna haplogroups coming with him?

" R1 " crops in paleolithic Italy and in modern " not equatorial " Africa, do you feel a subtility? The same group of modern African R1b pops in mesolithic Balkans and neolithic Ukraine / Russia. No subtility yet?
There were probably numerous R1 men venturing out of the Steppe, temporarily or permanently, over tens of thousands of years; each would have had different autosomal and mitochondrial DNA, depending on their unique chains of ancestry.
The divergence between African V88 and Eurasian V88 is date-estimated by yfull at 5,000 BC; my estimate incorporating STR diversity is slightly longer. Either way, it is indicative of a likely migration during the Neolithic.

Pip
20-10-18, 21:44
Yes its called multiple founder effects. R1b-Z2103 lost predominance everywhere after the Bronze Age, while it survive pretty good in Anatolia and Armenia. Thanks to mountaneous regions and large genetic diversity. Europe is a geographical bottleneck, while middle-east have more holes than swiss cheese.
This could be a vital point. Lost predominance sometimes means near-eradication. The most interesting branches are the ones that survived and re-developed, and these are often not the branches that initially proliferated, nor those for which ancient samples have been found.
The major branch of Z2103 today is an Eastern variety which most likely split from the Yamnayan European branch some time before its colonisation of the Danube basin took place. My suggestion is that it was most likely a minority component within Maykop, and possibly a later major source of the Hittites.

Silesian
20-10-18, 21:58
What do i know?

You know there is a difference in branching between Steppe [Yamnaya], Caucasus, R1b-Z2103 found in Armenian, Ossetians, Bashkirs?

Here is an interesting video @32 minutes the Yamnaya displaced, absorbed or wiped out other Steppe groups; Cucuteni-Trypillia was also nearby Yamnaya and might have suffered a similar fate. https://en.wikipedia.org/wiki/Cucuteni%E2%80%93Trypillia_culture

https://www.youtube.com/watch?v=fHdCuhYRHqo

halfalp
20-10-18, 23:10
You know there is a difference in branching between Steppe [Yamnaya], Caucasus, R1b-Z2103 found in Armenian, Ossetians, Bashkirs?

Here is an interesting video @32 minutes the Yamnaya displaced, absorbed or wiped out other Steppe groups; Cucuteni-Trypillia was also nearby Yamnaya and might have suffered a similar fate. https://en.wikipedia.org/wiki/Cucuteni%E2%80%93Trypillia_culture

https://www.youtube.com/watch?v=fHdCuhYRHqo

Branching doesn't matter, it only tell us prehistoric R1b-Z2103 from eastern europe and modern R1b-Z2103 from Armenia, are different. But they have ultimately a same origin and a same autosomal dna, same mtdna lineage. Pip point doesn't make sense, populations of Africa with R1b-V88 have also mtdna lineage like U5b that are ultimately european. Modern population have obviously different autosomal dna than ancient one.

halfalp
20-10-18, 23:15
There were probably numerous R1 men venturing out of the Steppe, temporarily or permanently, over tens of thousands of years; each would have had different autosomal and mitochondrial DNA, depending on their unique chains of ancestry.
The divergence between African V88 and Eurasian V88 is date-estimated by yfull at 5,000 BC; my estimate incorporating STR diversity is slightly longer. Either way, it is indicative of a likely migration during the Neolithic.

This is absolutely not true for europe before the neolithic... People tend to keep the same lineages and the same ancestry until populations with multiple lineage and mixtures from the middle-east enter in europe.

Pip
20-10-18, 23:55
This is absolutely not true for europe before the neolithic... People tend to keep the same lineages and the same ancestry until populations with multiple lineage and mixtures from the middle-east enter in europe.
On what basis do you arrive at this conclusion? Why would people from the Middle East mix, but not people from Europe? Why would people coming in from the Middle East then also make them start to mix? I'm not convinced.
Pre-Neolithic Northern European populations were in any case a mix of Western and Eastern Hunter Gatherer lineages. Yamnayans were, in any case, a mix of Eastern Hunter Gatherer and Iranian-like DNA. We're even told that some Paleolithics mixed with Neanderthals, leaving a small remnant of Neanderthal in present day European DNA.

Pip
21-10-18, 00:10
Branching doesn't matter, it only tell us prehistoric R1b-Z2103 from eastern europe and modern R1b-Z2103 from Armenia, are different. But they have ultimately a same origin and a same autosomal dna, same mtdna lineage. Pip point doesn't make sense, populations of Africa with R1b-V88 have also mtdna lineage like U5b that are ultimately european. Modern population have obviously different autosomal dna than ancient one.
Chalcolithic Z2103 samples from East Central Europe have both a strikingly different autosomal DNA and a different mitochondrial DNA mix to those from Russia. Russian samples have a significant Iranian-like component, and Balkan samples have a substantial input from indigenous European farmers.
African V88 with European mtDNA lineages shows the women (as well as the men) were prepared to move very long distances away from the Steppe during the Neolithic.

halfalp
21-10-18, 00:38
On what basis do you arrive at this conclusion? Why would people from the Middle East mix, but not people from Europe? Why would people coming in from the Middle East then also make them start to mix? I'm not convinced.
Pre-Neolithic Northern European populations were in any case a mix of Western and Eastern Hunter Gatherer lineages. Yamnayans were, in any case, a mix of Eastern Hunter Gatherer and Iranian-like DNA. We're even told that some Paleolithics mixed with Neanderthals, leaving a small remnant of Neanderthal in present day European DNA.

WHG and EHG is only the Input of ANE into WHG, it's not a big differentiation, and it lasted maybe even 10'000 millenia. Neolithic was Anatolia_N, Levante_N, Iran_N, Basal Eurasian ( if we imagine it existed somwhere in a very pure form ), CHG... HG's had a very low density population making a population distant of thousands of miles being the same autosomally, but farmers because of demographic explosions had every centuries to migrate somwhere else to encounter new people. The proof is, for 30'000 years, europe had only y-dna C1a,I and R with mtdna U2,u4,U5,U8,R,M ( a part M, all related in ancestry ) while neolithic in only couple of centuries had develop and absorb so many lineages ( mtdna H have like 95 subgroups ) and expanded so intensively, it is absolutely not the same story. Neanderthal addition doesn't change too much because it probably happened in a very small group and only resurface here and there with exchange of ancestry like Goyet -> El Miron.

halfalp
21-10-18, 00:43
Chalcolithic Z2103 samples from East Central Europe have both a strikingly different autosomal DNA and a different mitochondrial DNA mix to those from Russia. Russian samples have a significant Iranian-like component, and Balkan samples have a substantial input from indigenous European farmers.
African V88 with European mtDNA lineages shows the women (as well as the men) were prepared to move very long distances away from the Steppe during the Neolithic.

Ok i'm confused... that autosomal dna between an Ukrainian population and an Armenian one ( with the same haplogroup ) is confusing to you? Because i'm R1b-L2, but i dont have the same autosomal dna than an Iberian R1b-L2, it sounds pretty logic in term of genetic no? Do modern Indian people of R1a-Z93 have the same autosomal dna as the Bulgarian Chalcolithic R1a-Z93? Dont be silly. And your point about men and women having the behavioral reasoning to move long distance, i dont know what to thinka bout it.

Pip
21-10-18, 11:41
I was responding to your post Originally Posted by halfalp - Branching doesn't matter, it only tell us prehistoric R1b-Z2103 from eastern europe and modern R1b-Z2103 from Armenia, are different. But they have ultimately a same origin and a same autosomal dna, same mtdna lineage.

You seemed to be saying that Z2103 from Eastern Europe and Z2103 from Armenia had the same autosomal DNA and mtDNA.
That is my point - even within Eastern European Z2103, the autosomal DNA and mtDNA is strikingly different. By 3,000-2,800 BC, although these people had a male ancestor in common, they were living within families drawn from very different ancestral roots. I only raised this point as circumstantial evidence to indicate that their most recent common male ancestor might have been earlier than some imagine, and pre-dated the emergence of the Yamnayan culture.

We are straying off the point a bit, which is the modern distribution of Z2103. There seem to be two major strands of it - the European one and the Eastern one centred around the Caucasus/Northern Turkey. I am interested in whether the Eastern strand descends from a section of Caspian Yamnaya, Maykop and/or the Hittites. Is there any DNA evidence in relation to this?

Silesian
21-10-18, 12:21
I would have to disagree. Snp/str/autosomal branching + carbon dating samples are extremely informative. For example z-2103 and L-51 branch from L-23+, it would make sense to place L23+ L51+ Z2103+ in the same geographic region, str/snp/autosomal connection.
Until very recent we did not have a proper understanding of Caucasus and speculated on the main Yamanaya autosomal components.
With each new sample our understanding is improved[we can ask proper questions]. For example, we can see that Georgia @26000+/- years ago had different composition than CHG; all located in a region near Yamnaya and Maykop burials. While WHG represented by Villabruna cluster is a R1b branch.


To address this imbalance and to better understand the relationship of Europeans and Near Easterners, we report genome-wide data from two ~26 thousand year old individuals from Dzudzuana Cave in Georgia in the Caucasus from around the beginning of the LGM. Surprisingly, the Dzudzuana population was more closely related to early agriculturalists from western Anatolia ~8 thousand years ago than to the hunter-gatherers of the Caucasus from the same region of western Georgia of ~13-10 thousand years ago. [B]Most of the Dzudzuana population's ancestry was deeply related to the post-glacial western European hunter-gatherers of the 'Villabruna cluster', but it also had ancestry from a lineage that had separated from the great majority of non-African populations before they separated from each other, proving that such 'Basal Eurasians' were present in West Eurasia twice as early as previously recorded. We document major population turnover in the Near East after the time of Dzudzuana, showing that the highly differentiated Holocene populations of the region were formed by 'Ancient North Eurasian' admixture into the Caucasus and Iran and North African admixture into the Natufians of the Levant. We finally show that the Dzudzuana population contributed the

Dzudzuana Ice Age foragers: a different type of Caucasus hunter-gatherer (Lazaridis et al. 2018 preprint) (http://eurogenes.blogspot.com/2018/09/dzudzuana-ice-age-foragers-different.html)

halfalp
21-10-18, 12:35
I was responding to your post Originally Posted by halfalp - Branching doesn't matter, it only tell us prehistoric R1b-Z2103 from eastern europe and modern R1b-Z2103 from Armenia, are different. But they have ultimately a same origin and a same autosomal dna, same mtdna lineage.

You seemed to be saying that Z2103 from Eastern Europe and Z2103 from Armenia had the same autosomal DNA and mtDNA.
That is my point - even within Eastern European Z2103, the autosomal DNA and mtDNA is strikingly different. By 3,000-2,800 BC, although these people had a male ancestor in common, they were living within families drawn from very different ancestral roots. I only raised this point as circumstantial evidence to indicate that their most recent common male ancestor might have been earlier than some imagine, and pre-dated the emergence of the Yamnayan culture.

We are straying off the point a bit, which is the modern distribution of Z2103. There seem to be two major strands of it - the European one and the Eastern one centred around the Caucasus/Northern Turkey. I am interested in whether the Eastern strand descends from a section of Caspian Yamnaya, Maykop and/or the Hittites. Is there any DNA evidence in relation to this?



We are not off the point, it's exactly what i'm talking about. The autosomal dna of modern Caucasus/Nothern Turkey, cannot tell you anything about Yamnaya, Maykop and especially not the Hittites. See it's pretty simple, like i said the Pontic Zone is an Horseshoe, if we take from base the North Caucasus, we can roam and turn until arrive in the South Caucasus into a circum-pontic trail. R1b-Z2103* probably originate in that Horseshoe. But then, how do we know if R1b-Z2103 from Yamnaya is local and did not came from Caucasus/North Turkey? Or how do we know that modern Caucasus/North Turkey R1b-Z2103 didn't come from Eastern Europe? Well there is an easy reponse. The response is autosomal dna. But! Autosomal DNA from modern populations is completely irrelevent, Caucasus/North Turkey till Yamnaya and now, have recieved so much genetic impact from totally and distant admixtures than the genetic makeup is too complicate to see an origin for. However, prehistoric samples is absolutely not the same story, because every pop's that were not FIRST ultimately touch by the demic advance of EEF, like in Eastern Europe, were still almost in a pure form. Yamnaya had different ancestry, the big ancestry was EHG, the second CHG. They had mtdna lineage linked with neolithic europe ( EEF ) who didn't give them substantial farmer genetic makeup, then the only link from the south in Yamnaya is CHG. The problem with CHG in Yamnaya, it's too big, it doesn't fit the lineage that we know south of the caucasus or maikop with it, it is completely out of link with R1b-Z2103. If R1b-Z2103 would come from south, it would also bring minority of EEF and Iran_Neolithic, the genetic makeup of Yamnaya would be bigger, we only see suck link in the steppe area were EHG exchange dna with Maikop. Conclusion, ultimately, R1b-M269 could came to south caucasus from the Balkans sometimes in the mesolithic, get CHG ancestry and back migrate into eastern europe where they would encounter older R1b lineages and become dominant. But isn't it a complicate hypothesis? Looks like an hypothesis to fit modern ethnic agenda, like Armenian or Turk, Kurdish... Isn't?

Pip
21-10-18, 14:49
I wouldn't discard any data as 'completely irrelevant', nor see any data as 'pure'.

We can identify at least 3 very different types of historic Z2103 individuals:
1. Eastern Yamnaya - broadly Steppe mtDNA, autosomal Caucasian component, hardly any EEF or WHG
2. Ukraine - broadly Steppe mtDNA, autosomal WHG component, hardly any Caucasian or EEF
3. Balkan - mixed mtDNA, autosomal EEF component, hardly any Caucasian

The core DNA to all three looks Northern - perhaps Poland or Northern Ukraine?

Might Maykop and/or the Hittites have incorporated Z2103? Do we have any historic Maykop or Hittite DNA with which to compare to the other three types of Z2103 mentioned?

What data do we have to best help identify the source of the major branches of Z2103 found today in Turkey and the Caucasus? Based on the limited data that we have, could the principal sources for this Z2103 be the Hittites, Maykop, Russian Yamnaya, Ukrainian pre-Yamnaya and/or Balkan?

markod
21-10-18, 15:18
I wouldn't discard any data as 'completely irrelevant', nor see any data as 'pure'.

We can identify at least 3 very different types of historic Z2103 individuals:
1. Eastern Yamnaya - broadly Steppe mtDNA, autosomal Caucasian component, hardly any EEF or WHG
2. Ukraine - broadly Steppe mtDNA, autosomal WHG component, hardly any Caucasian or EEF
3. Balkan - mixed mtDNA, autosomal EEF component, hardly any Caucasian

The core DNA to all three looks Northern - perhaps Poland or Northern Ukraine?

Might Maykop and/or the Hittites have incorporated Z2103? Do we have any historic Maykop or Hittite DNA with which to compare to the other three types of Z2103 mentioned?

What data do we have to best help identify the source of the major branches of Z2103 found today in Turkey and the Caucasus? Based on the limited data that we have, could the principal sources for this Z2103 be the Hittites, Maykop, Russian Yamnaya, Ukrainian pre-Yamnaya and/or Balkan?

There's an Iron Age Hurrian with Z2103 or something under it IIRC.

I wouldn't put too much stock in autosomal DNA when it comes to the origin of haplogroups though. Lots of mixing in the metal ages.

Pip
22-10-18, 00:07
I have checked some samples that show a heavy Eastern Hunter Gatherer component (i) in Northern Maykop as far back as 3,500 BC, (ii) within 60 or so miles of Maykop city by 2,800 BC and (iii) on the Northern foothills of the Caucasus mountains by 2,750 BC. There are samples with heavy EHG all along the Western side of the Northern Caucasus.

The last of these three samples was identified as R1b-L23 (I would say most likely Z2103), and was located less than 200 miles from Armenia.

In the absence of other evidence, is it more likely that this (rather than a Balkan variety of Z2103) is the source of the Iron Age Hurrian Z2103 and the Armenian branch of modern Z2103?

halfalp
22-10-18, 01:46
Not sure to get you. Eastern Steppe was mostly WHG component through EHG than Caucasian. About Maikop, Maikop Steppe, North Caucasus HG's were autosomally mostly from the Steppe, the North Caucasus Eneolithic wich Maikop " Culture " is part of was CHG / Iran_Neolithic. Maikop Steppe was mostly R1b and Q1a2 while Maikop Culture was mostly J1,J2 and G, with some L if i recall, exactly like the southern counterpart Kura-Araxes. It is possible that R1b-Z2103 from Armenia was already at this time from the North Caucasus part and already more CHG shifted, then at late Yamnaya or Catacomb, they migrated through the Darial Pass with pretty much the same genetic makeup as Maikop. Catacomb Culture was also mostly R1b-Z2103 but i think the ancestry shows a more northern origin that Yamnaya, so maybe late Yamnaya people " run away " Catacomb people

Pip
22-10-18, 13:19
From the data I have seen:
1. Eastern Steppe had very little WHG in comparison to Ukrainian Steppe.
2. Northern Maykop had a greater EHG component than subsequent Z2103 Yamnaya did, suggesting a development area for Z2103 proto-Yamnaya and the addition of its Caucasian DNA principally within Maykop.
3. The southwards movement of Z2103 and other Northern Steppe DNA into Anatolia and Armenia seems to have happened gradually - (i) firstly into Northern Maykop (perhaps 3,500 BC), (ii) then branching away from Yamnaya and replacing Southern Maykop (perhaps 3,000 BC), (iii) then moving into the Western side of the Caucasus (perhaps 2,800 BC), (iv) then replacing Kura-Araxes (perhaps as a mixed ethnicity component or relative of Trialeti) in Armenia and North Eastern Anatolia (perhaps 2,400 BC).
4. As both the Yamnayans and the North Anatolian Hittites were probably Indo-European, I would suggest they most likely had some common paternal origins (including Z2103) within Northern Maykop, and that there was (and is) too little WHG in Anatolia for the proto-Hittites to have migrated into Anatolia from the Balkans.

Any thoughts?

halfalp
22-10-18, 17:21
From the data I have seen:
1. Eastern Steppe had very little WHG in comparison to Ukrainian Steppe.
2. Northern Maykop had a greater EHG component than subsequent Z2103 Yamnaya did, suggesting a development area for Z2103 proto-Yamnaya and the addition of its Caucasian DNA principally within Maykop.
3. The southwards movement of Z2103 and other Northern Steppe DNA into Anatolia and Armenia seems to have happened gradually - (i) firstly into Northern Maykop (perhaps 3,500 BC), (ii) then branching away from Yamnaya and replacing Southern Maykop (perhaps 3,000 BC), (iii) then moving into the Western side of the Caucasus (perhaps 2,800 BC), (iv) then replacing Kura-Araxes (perhaps as a mixed ethnicity component or relative of Trialeti) in Armenia and North Eastern Anatolia (perhaps 2,400 BC).
4. As both the Yamnayans and the North Anatolian Hittites were probably Indo-European, I would suggest they most likely had some common paternal origins (including Z2103) within Northern Maykop, and that there was (and is) too little WHG in Anatolia for the proto-Hittites to have migrated into Anatolia from the Balkans.

Any thoughts?

The thing is, the Maikop Proper samples that we have are from Late Maikop, not Early Maikop. So R1b-Z2103 have absolutely nothing to do with Maikop originally. But there is always a possibility that some steppe foragers neighboring Maikop actually take the whole cultural package of Maikop at some point and migrate slowely south of the Caucasus. If you look at the table of " Ancient genomes from Caucasus " you will see that Caucasus Yamnaya have more CHG than EHG, so no the CHG ancestry was gradual with a high admixture South-East ( Caucasus ) and a low admixture North-West ( CWC ), wich make sense if you think about it. The problem with the Caucasus hypothesis, is that we dont see a clear culture, there is no Yamnaya Anatolia or Maikop Anatolia, we dont see any culture to be link with a North -> South migration. So the migration probably happened with Boats or from the Balkans. I think the highest point of Z2103 in Anatolia is actually Sinop, wich could be consistant with a migration by sea.

Pip
22-10-18, 18:29
Yes, migration by sea is certainly a possibility, often ignored. But from where? Central Anatolia does not display sufficient WHG to suggest a migration from Ukraine or the Balkans, unless it occurred after WHG-heavy Steppe people had been decimated by Eastern invaders. More likely from Azov, perhaps in response to incursions from R1a-Z93-dominated groups?

Also, branching within Z2103 would suggest that the Anatolian varieties most likely arose in the East. Despite being more numerous perhaps at Sinop, it looks most diverse (both in terms of SNPs and STRs within SNPs) around Armenia.

As Hittite is also considered the earliest offshoot of Indo-European, this means it most likely predates the separation (and possibly even the preceding admixture) between R1b-M269 (Yamnayan and Bell Beaker) and R1a-M417 (Corded Ware) peoples, i.e. Anatolian sections would have branched off very early indeed or would have emerged from populations relatively untouched by R1a admixture (such as the Eastern Pontic).

halfalp
22-10-18, 19:16
Yes, migration by sea is certainly a possibility, often ignored. But from where? Central Anatolia does not display sufficient WHG to suggest a migration from Ukraine or the Balkans, unless it occurred after WHG-heavy Steppe people had been decimated by Eastern invaders. More likely from Azov, perhaps in response to incursions from R1a-Z93-dominated groups?

Also, branching within Z2103 would suggest that the Anatolian varieties most likely arose in the East. Despite being more numerous perhaps at Sinop, it looks most diverse (both in terms of SNPs and STRs within SNPs) around Armenia.

As Hittite is also considered the earliest offshoot of Indo-European, this means it most likely predates the separation (and possibly even the preceding admixture) between R1b-M269 (Yamnayan and Bell Beaker) and R1a-M417 (Corded Ware) peoples, i.e. Anatolian sections would have branched off very early indeed or would have emerged from populations relatively untouched by R1a admixture (such as the Eastern Pontic).

That or Coastal People from Ukraine run away some sort of flood or natural catastrophe and happened all around the black sea shores. Azov sea was one day some proper lands, and that time is not that far away in time. The Anatolian question is obviously the most tricky about all the IE question. I dont have any hypothesis in mind for now, obviously i lean more toward a northern origin, but we dont see some clear pattern of migration like CWC, BB, Sintashta, so the question remain.

Pip
25-10-18, 12:18
Looking at the branching of eastern/Armenian Z2103, I do not really see many early pan-Anatolian indicators, so I think R-PF7562 might be a better candidate for a Hittite/Luwian source. Modern eastern Z2103 looks most likely to be Caucasian leakage from proto-Yamnayan/Steppe Maykop that has remained fairly localised around Armenia, apart from a little southward spread into the Levant and Iraq.

Another question is what happened to Samaran/eastern Yamnayan Z2103. I cannot see any signs that it survived to any significant degree, and it looks instead as if it was its immediate ancestor (root Z2103) that survived in the West, although largely by being subsumed into R1b-L51 and R1a-M417 dominated populations.

berun
25-10-18, 15:33
I think the Z2103 in Hajji Firuz has been confirmed by others. It hasn't been dated directly however.

good this one
:)

when facts are against steppe homeland then there are so many objections... that my red and orange alarms are like sparks.

markod
25-10-18, 17:15
good this one
:)

when facts are against steppe homeland then there are so many objections... that my red and orange alarms are like sparks.

Wrong guy, I give slightly more chance to Armenia/Anatolia being confirmed in the future. I don't feel like arguing for the possibility anymore now that Reich and others are considering it.

halfalp
25-10-18, 19:57
I'm not sure to understand you both, are we talking about R1b-Z2103 or about PIE? origins? Because both are not the same.

halfalp
25-10-18, 20:10
I dont think the Reich Lab have any sample that matter. Why make 2 papers on the Caucasus, wich is basically the place we are targeting for a southern origin of both R1b-M269 and PIE ; and not give all the samples? I think they overestimated their Maikop paper, i think most of the linguists and archeologists that were working with them, were so autocenter to their Renfrew narrative, that they didn't think to the extra passionnate amateurish community, exactly like the Hajji Firuz sample. If one day they make a Shulaveri-Shomu or Leila-Tepe paper wich confirm a southern origin for both, they gonna have to quote numbers of amateurs.

halfalp
25-10-18, 20:24
Areni1: y-dna L1a - mtdna H2a1 / Khvalynsk: y-dna R1b1 - mtdna H2a1 / Sredni Stog: y-dna R1a - mtdna H2a1a.

Silesian
25-10-18, 22:26
..........Another question is what happened to Samaran/eastern Yamnayan Z2103.

They are related to Bashkirs

http://suyun.info/userfiles/bulletin/2017-7/z2103_r1b_paleo_dna.png


I cannot see any signs that it survived to any significant degree, and it looks instead as if it was its immediate ancestor (root Z2103) that survived in the West, although largely by being subsumed into R1b-L51 and R1a-M417 dominated populations.
Comparing apples with oranges.Neither R1b-L51 or R1a-M417 covered the geographical distance-time frame 5000YBP +- - R1b-Z2103/Yamnaya.
https://www.britannica.com/place/the-Steppe


The Steppe, belt of grassland (https://www.britannica.com/science/grassland) that extends some 5,000 miles (8,000 kilometres) from Hungary (https://www.britannica.com/place/Hungary) in the west through Ukraine (https://www.britannica.com/place/Ukraine) and Central Asia to Manchuria (https://www.britannica.com/place/Manchuria) in the east

markod
25-10-18, 23:13
I dont think the Reich Lab have any sample that matter. Why make 2 papers on the Caucasus, wich is basically the place we are targeting for a southern origin of both R1b-M269 and PIE ; and not give all the samples? I think they overestimated their Maikop paper, i think most of the linguists and archeologists that were working with them, were so autocenter to their Renfrew narrative, that they didn't think to the extra passionnate amateurish community, exactly like the Hajji Firuz sample. If one day they make a Shulaveri-Shomu or Leila-Tepe paper wich confirm a southern origin for both, they gonna have to quote numbers of amateurs.

The Maykop paper definitely weakens the southern hypothesis of PIE as formulated by Reich and Lazaridis. Hopefully they have some kind of strong evidence that will bring some closure one way or the other - if that's all they have that would be very disappointing.

It all hinges on samples from Anatolia, India & Greece anyway, and those are still few and far between.

halfalp
25-10-18, 23:20
The Maykop paper definitely weakens the southern hypothesis of PIE as formulated by Reich and Lazaridis. Hopefully they have some kind of strong evidence that will bring some closure one way or the other - if that's all they have that would be very disappointing.

It all hinges on samples from Anatolia, India & Greece anyway, and those are still few and far between.

My guess goes this way, we found R1b-V88 in mesolithic Balkans and Eastern Europe and modern Africa. It's an obvious means that north and south eurasia migrated in the other direction. So, sampling the good cultures, we can have some surprise. The thing about PIE and the Maikop paper, i suppose they take that stance " it's basically southern civilized guys and northern steppe foragers, the first probably have bring civilization to the second ". It's a bias like one another. Also, i've read by Carlos Quiles that apparently there is a future paper of Greece Neolithic with a Steppe Outlier in Neolithic Greece, that he link with a maybe Proto-Anatolian migration. Let's see the future.

Joey37
26-10-18, 02:56
I wouldn't believe anything Carlos Quiles wrote. His crackpot idea of the R1a people speaking Uralic invalidates any other work. Any Uralic population with R1a is a result of Slavic introgression; the haplogroup of the Uralics is N1c. Sanskrit and Lithuanian are too pure to have been the descendants of an Indo-European Uralic pidgin which would have been, in his theories, their parent. It is the Centum languages, the R1b tongues, that have the innovations. And the frequency of R1b-L51 drops precipitously east of the old Iron Curtain; it certainly has no part in spreading Indo-European tongues out of the steppe.

halfalp
26-10-18, 10:10
I wouldn't believe anything Carlos Quiles wrote. His crackpot idea of the R1a people speaking Uralic invalidates any other work. Any Uralic population with R1a is a result of Slavic introgression; the haplogroup of the Uralics is N1c. Sanskrit and Lithuanian are too pure to have been the descendants of an Indo-European Uralic pidgin which would have been, in his theories, their parent. It is the Centum languages, the R1b tongues, that have the innovations. And the frequency of R1b-L51 drops precipitously east of the old Iron Curtain; it certainly has no part in spreading Indo-European tongues out of the steppe.

I dont think 1 bias toward a certain hypothesis, invalidate a complete person. That's basically what people use as an argument to discredite a lot of people, without critical thinking.

Joey37
26-10-18, 14:52
I don't know, I guess you're right. But the idea of Corded Ware speaking Uralic makes no sense; there would be traces of it all the way into central Germany, and it would sort of invalidate the idea of Indo-European being a true language family, as some entire branches would be the result of language contact and replacement. I think Maciamo really illustrated this problem in his page on R1b-It has been suggested that Indo-European (IE) languages simply disseminated through contact, just like technologies, or because it was the language of a small elite and therefore its adoption conferred a certain perceived prestige. However people don't just change language like that because it sounds nicer or more prestigious. Even nowadays, with textbooks, dictionaries, compulsory language courses at school, private language schools for adults and multilingual TV programs, the majority of the people cannot become fluent in a completely foreign language, belonging to a different language family. The linguistic gap between pre-IE vernaculars and IE languages was about as big as between modern English and Chinese. English, Greek, Russian and Hindi are all related IE languages and therefore easier to learn for IE speakers than non-IE languages like Chinese, Arabic or Hungarian. From a linguistic point of view, only a wide-scale migration of IE speakers could explain the thorough adoption of IE languages in Western Europe - leaving only Basque as a remnant of the Neolithic languages". This speaks against demic diffusion; if R1a did not originally speak Indo-European languages, then they picked them up in the northern fringes of Yamna when it was still just a small community of forest-steppe people. I do believe, though, that R1b-Z2103 did play a major part in spreading Indo-European tongues; the distribution of this clade is much more widespread in Eurasia than the R1b-L51 descended clades.

Pip
26-10-18, 18:43
From the data l have seen, the modern distribution of Z2103 appears to fall into two basic strands - an Armenian offshoot of Northern Maykop and a Central European branch. If there are any other modern branches that do not fall within these two strands, what are their distinguishing SNPs or STR features?

halfalp
26-10-18, 19:16
I don't know, I guess you're right. But the idea of Corded Ware speaking Uralic makes no sense; there would be traces of it all the way into central Germany, and it would sort of invalidate the idea of Indo-European being a true language family, as some entire branches would be the result of language contact and replacement. I think Maciamo really illustrated this problem in his page on R1b-It has been suggested that Indo-European (IE) languages simply disseminated through contact, just like technologies, or because it was the language of a small elite and therefore its adoption conferred a certain perceived prestige. However people don't just change language like that because it sounds nicer or more prestigious. Even nowadays, with textbooks, dictionaries, compulsory language courses at school, private language schools for adults and multilingual TV programs, the majority of the people cannot become fluent in a completely foreign language, belonging to a different language family. The linguistic gap between pre-IE vernaculars and IE languages was about as big as between modern English and Chinese. English, Greek, Russian and Hindi are all related IE languages and therefore easier to learn for IE speakers than non-IE languages like Chinese, Arabic or Hungarian. From a linguistic point of view, only a wide-scale migration of IE speakers could explain the thorough adoption of IE languages in Western Europe - leaving only Basque as a remnant of the Neolithic languages". This speaks against demic diffusion; if R1a did not originally speak Indo-European languages, then they picked them up in the northern fringes of Yamna when it was still just a small community of forest-steppe people. I do believe, though, that R1b-Z2103 did play a major part in spreading Indo-European tongues; the distribution of this clade is much more widespread in Eurasia than the R1b-L51 descended clades.


Thing is, it's like Flat Earth. People who really are interested in this reality, you never gonna beat them, they gonna have hundred of arguments over you. The same goes for Carlos or Olympus Mons for exemple, or a lot of people, you can't really have a word because they know their " bias ", at the end of the day they might be wrong or right, or wrong in a way and right in another. I link Uralic with N1c, like i link IE with R1b, but maybe there is some subtilities that i dont think of, even with my big memory.

Aaron1981
27-10-18, 01:43
I wouldn't believe anything Carlos Quiles wrote. His crackpot idea of the R1a people speaking Uralic invalidates any other work. Any Uralic population with R1a is a result of Slavic introgression; the haplogroup of the Uralics is N1c. Sanskrit and Lithuanian are too pure to have been the descendants of an Indo-European Uralic pidgin which would have been, in his theories, their parent. It is the Centum languages, the R1b tongues, that have the innovations. And the frequency of R1b-L51 drops precipitously east of the old Iron Curtain; it certainly has no part in spreading Indo-European tongues out of the steppe.

You haven't really thought this through have you? There are only about 600 years separating both Z2103 and L51 from their immediate parent, and L51+ hasn't been found any earlier than ~2500 BC.

Pip
28-10-18, 13:03
I'm still interested in whether there is any evidence for a significant modern branch of Z2103 (Bashkir or otherwise) that does not appear Central European or Maykop/Armenian in origin. If we cannot find any, do we not have to provisionally conclude that Eastern Yamnayan male DNA predominantly died out? This is especially the case, as both West European L51 and Central European branches of Z2103 appear to have split away from Eastern (Russian/Caucasian) branches several hundred years before the Yamnayan culture arose.

Pip
02-11-18, 00:22
The Western branch of today's Z2103 looks to have been formative primarily in North Eastern Europe, near to the fringe of the area occupied by Neolithic farming communities:
1. The core autosomal components of its ancient samples are Eastern Hunter Gatherer and Early European Farmer, with relatively small amounts of Western and Caucasian Hunter Gatherer.
2. Its SNP branches and the STR variance within those branches coalesce back to an estimated origin point somewhere near Poland.

Data indicates that Western Z2103's most successful branches most likely moved Southwards into the Balkans, Italy, Ukraine (and perhaps even Northern Spain) during the late 4th millennium BC. Its modern distribution has little or no correlation with the autosomal DNA proportions found in early Steppe populations, which are unlikely to be core Z2103 as their WHG component is far too high for this. The surviving branches of Western Z2103 look pretty much a balanced mix of EHG and EEF, suggesting a collaboration between these two ancestral populations that split away and evacuated East Central Europe approximately when the Yamnayans began to colonise it, and at least several centuries before the major Bell Beaker expansions took place.

Branches of Z2103 that remained in Northern Europe or settled in Central Ukraine look to have been subsumed into Corded Ware and Yamnayan populations before these too were mostly eliminated by adverse conditions or other groups.

Brother clades Western Z2103 and R1b-L51 seem to demonstrate a pretty much identical estimated development pattern until around the mid to early 4th millennium BC, when Z2103 took more Southerly paths and L51 moved into Northern France.

Z2103 looks perhaps to have been pushed to the Southern and South Eastern fringes of Europe by expansions from R1a-dominated and then L51-dominated populations.

Are there any more suggestions on how Z2103 got to where it is now most heavily distributed?

KirmanjeDersim
12-11-18, 18:59
Hello,

I am Kurdish, Zaza-speakers of Dersm region (Tunceli in Turkish). My Y-DNA is R1b-Z2103 > PH2731 > PH655.

My question is my Y-DNA is a Armenian sub-clades ? Actually I don’t understand, many administrators of Armenia DNA project say that my ancestry is Armenian.

Actually some my Y-DNA match say that they are Armenian of Dersim.

PS : Sorry for my English.

Pip
15-11-18, 01:35
Hello,

I am Kurdish, Zaza-speakers of Ders�m region (Tunceli in Turkish). My Y-DNA is R1b-Z2103 > PH2731 > PH655.

My question is my Y-DNA is a Armenian sub-clades ? Actually I don’t understand, many administrators of Armenia DNA project say that my ancestry is Armenian.

Actually some my Y-DNA match say that they are Armenian of Dersim.

PS : Sorry for my English.
Yes, your y-DNA looks Armenian to me, going back to at least the early 3rd millennium BC.

KirmanjeDersim
20-11-18, 19:24
Is it possible that my Y-DNA is of Iranian and not Armenian or even Anatolian origin? (Hittite)

Lenab
21-11-18, 17:47
Is it possible that my Y-DNA is of Iranian and not Armenian or even Anatolian origin? (Hittite)
No R1b is from the Anatolian/Armenian origin. There is a video on YT called the Armenian dna project stating it. What's your origin?

KirmanjeDersim
30-12-18, 00:37
I am Kurdish Zazaki-speakers of Dersim region (in Turkish : Tunceli) in Eastern Anatolia. In Family Tree DNA, I realised Y-DNA test and I have some Armenian match but they Armenian are Dersim Armenian. Two of my branch make Big-Y and we are new Mutation now we are R1b-Z2103 > L584 > PH1639 > PH2731 > PH655 > PH1731 > PH3425 > PH3610.
But two PH655 in Armenia make Big Y and they mutation are
PH2731 > PH655 > PH1731 > PH3425.
And datation is 1650 years ago, so I think that we separated from this date. But I don’t know.

dosas
07-02-19, 09:53
Hello, I am new to this forum. I am currently waiting my Big-Y700 results, but I have done the Y-111 test before that. Feeding those results into Nevgen it gives me R1b Z2103>L584> PF7580. My paternal lineage is Greek Thracian afaik from Ivaylovgrad (currently Bulgaria).

Pip
12-05-19, 01:30
There has been a lot of new data coming out on R1b-Z2103 regarding both its SNP phylogeny and STR diversity. ftDNA now publishes details on 278 samples. The data is now of sufficient volume and precision to enable estimates to be drawn with a reasonable degree of confidence.

My calculations now estimate an earlier TMRCA for today's Z2103 lineages (4,700 BC, compared to yfull's 3,500 BC) - three of its basal branches (PF331, Y13369 and Y4364) coalesce to estimated origin points south of the Caucasus, as do their basal sub-branches. The fourth basal branch (Z8131) also has a basal sub-branch (PH4902) coalescing to the Southern Caucasus. The hypothesis that extant Z2103's earliest development occurred in the Steppe, and that it only moved south of the Caucasus with Indo-European migrations in the mid 2nd millennium BC accordingly does not look credible.

Perhaps Z8131 expanding into the Steppe over the 5th and 4th millennia was one of the main sources of the increasing CHG component that arose there over this period?

halfalp
12-05-19, 09:53
There has been a lot of new data coming out on R1b-Z2103 regarding both its SNP phylogeny and STR diversity. ftDNA now publishes details on 278 samples. The data is now of sufficient volume and precision to enable estimates to be drawn with a reasonable degree of confidence.

My calculations now estimate an earlier TMRCA for today's Z2103 lineages (4,700 BC, compared to yfull's 3,500 BC) - three of its basal branches (PF331, Y13369 and Y4364) coalesce to estimated origin points south of the Caucasus, as do their basal sub-branches. The fourth basal branch (Z8131) also has a basal sub-branch (PH4902) coalescing to the Southern Caucasus. The hypothesis that extant Z2103's earliest development occurred in the Steppe, and that it only moved south of the Caucasus with Indo-European migrations in the mid 2nd millennium BC accordingly does not look credible.

Perhaps Z8131 expanding into the Steppe over the 5th and 4th millennia was one of the main sources of the increasing CHG component that arose there over this period?

That's comparing Steppe R1b-Z2103* probably dead with modern R1b-Z2103 south of the caucasus. Obviously, they will not show as coming from somewhere else than south of caucasus, that should be obvious. After coming from Steppe, all those " basal branches " had a founder effect south of caucasus. Only ancient dna can explain you where it was, not modern dna.

Pip
12-05-19, 12:19
That's comparing Steppe R1b-Z2103* probably dead with modern R1b-Z2103 south of the caucasus. Obviously, they will not show as coming from somewhere else than south of caucasus, that should be obvious. After coming from Steppe, all those " basal branches " had a founder effect south of caucasus. Only ancient dna can explain you where it was, not modern dna.

I'm not so interested in dead lines, only the branch from which modern Z2103 developed. From SNPs, yfull estimates this modern Z2103 branched four ways in 3,500 BC; from STRs, I estimate it started branching sooner (4,700 BC).

You say "after coming from Steppe", as if that were a given:
1. It is important to define what would have come from the Steppe - do you mean Z2103 in formation, fully formed Z2103 or various subclades of Z2103? (These could be separated by thousands of years)
2. Z2103's uncle clade PF7562 also seems to coalesce south of the Caucasus, indicating a good possibility that the whole family of living M269 derived from that region.
3. When would the founder effect of the South of the Caucasus basal branches have arisen? STR variability suggests this occurred very early (5th millennium BC, or 3,500 BC per yfull), certainly nowhere near the mid 2nd millennium BC when autosomal DNA suggests that EHG-heavy Steppe DNA arrived there.

Ancient DNA explains where ancient people lived, but we have no way of knowing whether any of these people were our ancestors. People actually ancestral to us might have lived somewhere completely different, so ancient DNA (like modern DNA) is just a guide, not the magic bullet that many seem to imagine.

Modern DNA can also act as a guide to where ancient DNA was. We used to realise that, but have lazily forgotten it. For instance, there are many precise branches of Z2103 (including at the basal level) within which the greatest range is exhibited in modern samples from Armenia. The quantity and precision of the data reflecting this is now sufficient to allow us to say with some confidence that this is not coincidence. The other explanations are that (i) living Z2103 developed somewhere other than Armenia and all remained together without any leakage anywhere else for thousands of years before migrating en masse to Armenia (highly implausible, especially as we know that Yamnayan early Z2103 was highly mobile and expansionary), or (ii) living Z2103 began developing in the general vicinity of Armenia and that only one basal branch of it spread at an early stage into the Steppe and beyond). If scenario (ii) seems to fit most plausibly with the data, then it seems reasonable to hypothesise that it brought autosomal CHG with it as it spread.

Piro Ilir
12-05-19, 18:50
Z2103 seems related mainly with Armenians, and secondly with Albanians.
It would be good to know which one of Z2103 subbranches (Z2106, L584, L277.1) is the Armenian one.

halfalp
12-05-19, 21:49
I'm not so interested in dead lines, only the branch from which modern Z2103 developed. From SNPs, yfull estimates this modern Z2103 branched four ways in 3,500 BC; from STRs, I estimate it started branching sooner (4,700 BC).

You say "after coming from Steppe", as if that were a given:
1. It is important to define what would have come from the Steppe - do you mean Z2103 in formation, fully formed Z2103 or various subclades of Z2103? (These could be separated by thousands of years)
2. Z2103's uncle clade PF7562 also seems to coalesce south of the Caucasus, indicating a good possibility that the whole family of living M269 derived from that region.
3. When would the founder effect of the South of the Caucasus basal branches have arisen? STR variability suggests this occurred very early (5th millennium BC, or 3,500 BC per yfull), certainly nowhere near the mid 2nd millennium BC when autosomal DNA suggests that EHG-heavy Steppe DNA arrived there.

Ancient DNA explains where ancient people lived, but we have no way of knowing whether any of these people were our ancestors. People actually ancestral to us might have lived somewhere completely different, so ancient DNA (like modern DNA) is just a guide, not the magic bullet that many seem to imagine.

Modern DNA can also act as a guide to where ancient DNA was. We used to realise that, but have lazily forgotten it. For instance, there are many precise branches of Z2103 (including at the basal level) within which the greatest range is exhibited in modern samples from Armenia. The quantity and precision of the data reflecting this is now sufficient to allow us to say with some confidence that this is not coincidence. The other explanations are that (i) living Z2103 developed somewhere other than Armenia and all remained together without any leakage anywhere else for thousands of years before migrating en masse to Armenia (highly implausible, especially as we know that Yamnayan early Z2103 was highly mobile and expansionary), or (ii) living Z2103 began developing in the general vicinity of Armenia and that only one basal branch of it spread at an early stage into the Steppe and beyond). If scenario (ii) seems to fit most plausibly with the data, then it seems reasonable to hypothesise that it brought autosomal CHG with it as it spread.

I think you clearly understimate what happens between, let's sayin, Yamnaya and 2K19. Everything nowadays have nothing to do with what happens to those times. You are saying you are not interested in Z2103* but then you take the conclusion over modern samples to dictate were it should be at the time or even before Z2103*. This methodology gonna bring you nowhere sadly.

There is an eazy possibility to your questioning of modern Z2103 branches. Only the one south of caucasus subsisted, while the one in the Pontic Steppe were replaced by R1a.

Also the term " basal branches " is totally missleading the entire results. There is Z2103 in modern Ukraine, Poland and Eastern Europe in general, we should assume with high possibilities, that those guys are the descendants of the Z2103* found in prehistoric Pontic Steppe. Therefore, the only thing that differentiate the Z2103 from Eastern Europe and the one from Armenia, is an older founder effect and TMRCA for the one in Armenia. Exactly the same mistakes were done about Kurdish samples of " basal branches of R1a ".

Then you have the mainstream hypothesis that both R1a and R1b transmigrated from Middle-East to Europe and from Europe to Middle-East from Epipaleolithic to Bronze Age. But more ancient dnas will come, more this idea gonna be rejected.

Pip
12-05-19, 23:03
Z2103 seems related mainly with Armenians, and secondly with Albanians.
It would be good to know which one of Z2103 subbranches (Z2106, L584, L277.1) is the Armenian one.
I would say most likely both of the latter two, and quite possibly Z2106 as well.

Pip
13-05-19, 00:08
I think you clearly understimate what happens between, let's sayin, Yamnaya and 2K19.
I've made no estimate of what has happened between Yamnaya and now.
On the contrary, it is the idea that Z2103 emerged on the Steppe and separated into lots of different branches (nearly all of which stuck together for thousands of years before migrating en masse to Armenia) that is based on a highly implausible underestimate of how Yamnaya and its descendants would have developed and spread around.


Everything nowadays have nothing to do with what happens to those times.
Of course it does. People with Z2103 nowadays are only Z2103 nowadays because their joint ancestor acquired the Z2103 mutation in those times.


You are saying you are not interested in Z2103*
No, I said that I wasn't so interested in dead lines.


but then you take the conclusion over modern samples to dictate were it should be at the time or even before Z2103*.
I'm not dictating anything. I am merely noting where SNP and STR calculations indicate is the most likely point of common origin.


This methodology gonna bring you nowhere sadly. There is an eazy possibility to your questioning of modern Z2103 branches. Only the one south of caucasus subsisted, while the one in the Pontic Steppe were replaced by R1a.
But this isn't what happened. There were lots of branches that subsisted South of the Caucasus, not just one. And the one branch in the Pontic Steppe was not entirely replaced by R1a, but still subsists today just as numerously (if not more so) than the Armenian branches.


Also the term " basal branches " is totally missleading the entire results. There is Z2103 in modern Ukraine, Poland and Eastern Europe in general, we should assume with high possibilities, that those guys are the descendants of the Z2103* found in prehistoric Pontic Steppe. Therefore, the only thing that differentiate the Z2103 from Eastern Europe and the one from Armenia, is an older founder effect and TMRCA for the one in Armenia.
There is nothing misleading about it.
Firstly, any Z2103* found in the prehistoric Pontic Steppe is pretty unlikely to be an ancestor of today's Eastern European Z2103 people, as (i) I estimate that Z2103 had already separated into its surviving subclades earlier on, and (ii) nearly all such ancient Z2103 lineages would have died out (Z2108 being the only steppic looking one).
Secondly, you are not going back far enough. The question is where did the basal development of living Z2103 occur? If there is one half-steppic/half-Southern Caucasus basal branch and three wholly Southern Caucasus basal branches, the balance surely tips in favour of their most recent common ancestor having been South Caucasian.

That is not to say that Z2103 was not in the Steppe from early on. There might well have been lots of Z2103 lineages in the early Steppe that became extinct; and one from an estimated 4,200 BC (Z8131) that survived and flourished there.

halfalp
13-05-19, 06:15
I've made no estimate of what has happened between Yamnaya and now.
On the contrary, it is the idea that Z2103 emerged on the Steppe and separated into lots of different branches (nearly all of which stuck together for thousands of years before migrating en masse to Armenia) that is based on a highly implausible underestimate of how Yamnaya and its descendants would have developed and spread around.


Of course it does. People with Z2103 nowadays are only Z2103 nowadays because their joint ancestor acquired the Z2103 mutation in those times.


No, I said that I wasn't so interested in dead lines.


I'm not dictating anything. I am merely noting where SNP and STR calculations indicate is the most likely point of common origin.


But this isn't what happened. There were lots of branches that subsisted South of the Caucasus, not just one. And the one branch in the Pontic Steppe was not entirely replaced by R1a, but still subsists today just as numerously (if not more so) than the Armenian branches.


There is nothing misleading about it.
Firstly, any Z2103* found in the prehistoric Pontic Steppe is pretty unlikely to be an ancestor of today's Eastern European Z2103 people, as (i) I estimate that Z2103 had already separated into its surviving subclades earlier on, and (ii) nearly all such ancient Z2103 lineages would have died out (Z2108 being the only steppic looking one).
Secondly, you are not going back far enough. The question is where did the basal development of living Z2103 occur? If there is one half-steppic/half-Southern Caucasus basal branch and three wholly Southern Caucasus basal branches, the balance surely tips in favour of their most recent common ancestor having been South Caucasian.

That is not to say that Z2103 was not in the Steppe from early on. There might well have been lots of Z2103 lineages in the early Steppe that became extinct; and one from an estimated 4,200 BC (Z8131) that survived and flourished there.

If you say so i guess. Your window for the formation of Z2103 in Armenia is contemporary with Areni in Armenia ( No R1b ), then the same contemporary population founded Maykop later ( No R1b ). We are then searching a population that never exister for now in ancient DNA, maybe one day it will show.

But i stay on my position that your estimations are a wrong methodology to explain ancient dna. Founder effects can litterally create a new origin for a lineage, V88 in Africa is a good exemple.

Pip
14-05-19, 09:50
If you say so i guess. Your window for the formation of Z2103 in Armenia is contemporary with Areni in Armenia ( No R1b ), then the same contemporary population founded Maykop later ( No R1b ). We are then searching a population that never exister for now in ancient DNA, maybe one day it will show.
But i stay on my position that your estimations are a wrong methodology to explain ancient dna. Founder effects can litterally create a new origin for a lineage, V88 in Africa is a good exemple.
This illustrates the problem with modern genetics, which appears to have gone backwards.
Three male skeletons (yDNA L, probably from the same family) are found in a single cave (Areni), and because they are not R1b, the conclusion is reached that R1b didn't exist South of the Caucasus. Presumably, this means there was no J, G or E South of the Caucasus either.
I am aware of 7 identified Maykop samples, one of which was R1 (presumably R1b, as there were several other North Caucasus R1b samples dated to around that time). There are also several other Maykop samples with unidentified Y DNA, and all of samples' yDNA readings are imprecise.
Sweeping conclusions have been reached on the basis of very little information.

Piro Ilir
14-05-19, 16:58
This illustrates the problem with modern genetics, which appears to have gone backwards.
Three male skeletons (yDNA L, probably from the same family) are found in a single cave (Areni), and because they are not R1b, the conclusion is reached that R1b didn't exist South of the Caucasus. Presumably, this means there was no J, G or E South of the Caucasus either.
I am aware of 7 identified Maykop samples, one of which was R1 (presumably R1b, as there were several other North Caucasus R1b samples dated to around that time). There are also several other Maykop samples with unidentified Y DNA, and all of samples' yDNA readings are imprecise.
Sweeping conclusions have been reached on the basis of very little information.

Interesting facts. From what period are the three samples from Areni? Is that calcolithic?

Even the Maykop samples are intriguing. Seems that my presumption about Maykop, is coming true. Is there any chance to know what R1_branch that sample might be?
I predicted that Maykop culture might have been the homeland of proto Hittites.

Pip
14-05-19, 19:18
Interesting facts. From what period are the three samples from Areni? Is that calcolithic?
Even the Maykop samples are intriguing. Seems that my presumption about Maykop, is coming true. Is there any chance to know what R1_branch that sample might be?
I predicted that Maykop culture might have been the homeland of proto Hittites.
If I remember correctly, the Areni samples are early 5th millennium BC Chalcolithic. As with other early R1 samples, no precise subclade is identified - most likely R1b-V88 or R1b-Z2103, as the other early Northern Caucasus R samples are.
I believe what distinguishes the R1 sample from other Maykop is its sizeable Anatolian component, which is suggestive of admixture from the South West.

halfalp
14-05-19, 20:32
This illustrates the problem with modern genetics, which appears to have gone backwards.
Three male skeletons (yDNA L, probably from the same family) are found in a single cave (Areni), and because they are not R1b, the conclusion is reached that R1b didn't exist South of the Caucasus. Presumably, this means there was no J, G or E South of the Caucasus either.
I am aware of 7 identified Maykop samples, one of which was R1 (presumably R1b, as there were several other North Caucasus R1b samples dated to around that time). There are also several other Maykop samples with unidentified Y DNA, and all of samples' yDNA readings are imprecise.
Sweeping conclusions have been reached on the basis of very little information.

I think it's pretty clear, it's not about R1b, it's about ancient dna. Nothing can explain history better than ancient dna. We will never found the perfect proxys. Modern branches are too diversified and had too much founder effect to be relevant of any kind. If a south caucasus R1b-Z2103 needs to exist in time, it will being found at some point. But if we follow your modern basal branches hypothesis, we should not just found R1b-Z2103* in prehistoric Armenia but also L23* and M269*.

halfalp
14-05-19, 20:37
This illustrates the problem with modern genetics, which appears to have gone backwards.
Three male skeletons (yDNA L, probably from the same family) are found in a single cave (Areni), and because they are not R1b, the conclusion is reached that R1b didn't exist South of the Caucasus. Presumably, this means there was no J, G or E South of the Caucasus either.
I am aware of 7 identified Maykop samples, one of which was R1 (presumably R1b, as there were several other North Caucasus R1b samples dated to around that time). There are also several other Maykop samples with unidentified Y DNA, and all of samples' yDNA readings are imprecise.
Sweeping conclusions have been reached on the basis of very little information.

The R1 individual you are mentionning is not from Maykop Proper, but Steppe Maykop, whatever it means on an archeological stand point. And it makes sense for R1 to be in North Caucasus at this point in time, because it's everywhere in Eastern Europe for milleniums prior to it. Also there is a Q1a2 individual alongside the R1 one, why not assume then too that this Q1a2 comes from chalcolithic Armenia.

Piro Ilir
14-05-19, 20:59
If I remember correctly, the Areni samples are early 5th millennium BC Chalcolithic. As with other early R1 samples, no precise subclade is identified - most likely R1b-V88 or R1b-Z2103, as the other early Northern Caucasus R samples are.
I believe what distinguishes the R1 sample from other Maykop is its sizeable Anatolian component, which is suggestive of admixture from the South West.

I bet for R-PF7562. Z2103 is too steppe related I think. V88 is also too southern related with Egypt and Africa.

IE Anatolian speakers were not steppe related (autosomal), hence my best bet go for PF7562, which is an early bronze age split from its branch sister L23. The best place of this split is northern Caucasus.

Ygorcs
14-05-19, 22:58
The R1 individual you are mentionning is not from Maykop Proper, but Steppe Maykop, whatever it means on an archeological stand point. And it makes sense for R1 to be in North Caucasus at this point in time, because it's everywhere in Eastern Europe for milleniums prior to it. Also there is a Q1a2 individual alongside the R1 one, why not assume then too that this Q1a2 comes from chalcolithic Armenia.

Besides, Steppe Maykop had a much more "eastern" and quite probably Central Asian Steppe autosomal makeup, very distinct from that of "truly" Caucasian Maykop. I wouldn't analyze these two cultures and peoples together in terms of genetic origins and possibly even ethnic and linguistic identity.

halfalp
15-05-19, 06:28
Besides, Steppe Maykop had a much more "eastern" and quite probably Central Asian Steppe autosomal makeup, very distinct from that of "truly" Caucasian Maykop. I wouldn't analyze these two cultures and peoples together in terms of genetic origins and possibly even ethnic and linguistic identity.

I think the guys from the paper clearly said that there was a distinction between Maykop and Maykop Steppe ( wich is why the distinction " Steppe Maykop " ) both in ancestry and y-dna. As for ethnic and linguistic, i'm not sure, mtdna of Maykop Steppe is clearly linked with the southern fellows and i've read here and there that some Steppe Maykop people, like Progress had cultural links towards south such as similar kurgans as leyla-tepe.

But as far as y-dna R1, i think it's almost clear that it came from eastern europe, even Narasimhan apparently had after few C14 fails to redate the Z2103 individual from Hajji Firuz something like MLBA. Everything tends to show that prior 3000 BC, there was no R1 south of caucasus, but that after, something triggered their movement.

Pip
16-05-19, 09:50
I think it's pretty clear, it's not about R1b, it's about ancient dna. Nothing can explain history better than ancient dna. We will never found the perfect proxys. Modern branches are too diversified and had too much founder effect to be relevant of any kind. If a south caucasus R1b-Z2103 needs to exist in time, it will being found at some point. But if we follow your modern basal branches hypothesis, we should not just found R1b-Z2103* in prehistoric Armenia but also L23* and M269*.
This thread is about the modern distribution of R1b, not ancient DNA, much of which is the history of dead-end lineages with little or no relevance to our ancestry.

If Z2103 South of the Caucasus is due to recent founder effects, then there must have been lots of different founders whose ancestors branched apart thousands of years beforehand, and who then all coincidentally ended up in the same small place (Armenia), leaving no trace of anyone related to them in the Steppe where they supposedly originated.

M269 formed over several thousand years. There are probably remnants of thousands of M269 people dotted all over the place. Whether they will all be found, got tested, be successfully read for yDNA and their results published is another matter. The idea that early M269 people all stuck together in one specific location and mated with women with similar autosomal profiles is an illusion.

Of course, if a single early M269 sample is found on the Steppe, it will be a case of 'proof that M269 was a Steppe lineage'. If found elsewhere, it will be ignored as 'noise' or an 'outlier'.

Pip
16-05-19, 16:29
The R1 individual you are mentionning is not from Maykop Proper, but Steppe Maykop, whatever it means on an archeological stand point. And it makes sense for R1 to be in North Caucasus at this point in time, because it's everywhere in Eastern Europe for milleniums prior to it. Also there is a Q1a2 individual alongside the R1 one, why not assume then too that this Q1a2 comes from chalcolithic Armenia.
The argument 'it's not proper' is often used in genetics debate. Bell Beakers without Steppe DNA are written off as not proper Bell Beaker. M269 with few SNP reads are ignored as not real M269. It's usually a symptom of denial.

If R1 was in the Northern Caucasus, why adamantly presume that no one bearing it could ever have crossed onto the other side of the mountains?

The reason why we should not assume that Q1a2 comes from Chalcolithic Armenia is simple - the phylogenic and STR variance data do not support it.

halfalp
16-05-19, 22:43
The argument 'it's not proper' is often used in genetics debate. Bell Beakers without Steppe DNA are written off as not proper Bell Beaker. M269 with few SNP reads are ignored as not real M269. It's usually a symptom of denial.

If R1 was in the Northern Caucasus, why adamantly presume that no one bearing it could ever have crossed onto the other side of the mountains?

The reason why we should not assume that Q1a2 comes from Chalcolithic Armenia is simple - the phylogenic and STR variance data do not support it.

Because,

1) How would R1b-Z2103 ( if it was born and came from South Caucasus ) and with a majority of Farmer mtdna, be that high in EHG ancestry?

2) R1 did cross Caucasus but in Chalcolithic-EBA transition when it became by founder effect and conquest one of the new lineage of late Kura-Araxes.

3) Q1a2 is absent of South Caucasus modern dna, why would STR variance change anything about the deduction of its origin not there.

4) While R1b-Z2103 had multiple founder effects in south caucasus for 5000 years, you are talking of denial, but this is a huge one. Absolutely all scientists using your hypothesis and statistics using modern dna were wrong for now.

Your same hypothesis made scientists be clear in 2011 that R1b-M269 came in europe with LBK. Your same hypothesis made scientists be clear that R1a-M417 had a diversification origin in Iran.

And as i said it, if M269, L23, Z2103 were south of the caucasus, it will show it at some point in a future study. Why trying to get a response only by modern dna conclusions?

halfalp
16-05-19, 22:56
This thread is about the modern distribution of R1b, not ancient DNA, much of which is the history of dead-end lineages with little or no relevance to our ancestry.

If Z2103 South of the Caucasus is due to recent founder effects, then there must have been lots of different founders whose ancestors branched apart thousands of years beforehand, and who then all coincidentally ended up in the same small place (Armenia), leaving no trace of anyone related to them in the Steppe where they supposedly originated.

M269 formed over several thousand years. There are probably remnants of thousands of M269 people dotted all over the place. Whether they will all be found, got tested, be successfully read for yDNA and their results published is another matter. The idea that early M269 people all stuck together in one specific location and mated with women with similar autosomal profiles is an illusion.

Of course, if a single early M269 sample is found on the Steppe, it will be a case of 'proof that M269 was a Steppe lineage'. If found elsewhere, it will be ignored as 'noise' or an 'outlier'.

You are starting to show impatience over the topic like " you want it to be south of the caucasus ". Bias can be hard times. If nobody would have ramble on the Hajji Firuz R1b-Z2103, we probably would have accepted the idea that this was the " ancestor of yamnaya guys " like Markod said impulsively and with joy at the preprint of the paper, wich also show'd his bias.

I dont believe to probabilities and statistics in ancient and modern dna. My point is, i believe prehistoric steppe Z2103 are older than any living Z2103. All modern southern caucasus Z2103 were born in south caucasus post-chalcolithic, mountaneous region are creating lots of founder effects very fast, as for very populous regions. India show the same kind of Z93 founder effects as Z2103 south of caucasus. Being an outlier depend on the ancestral component % according to the whole population analyzed, so it's not relevant with lineage.

Thing is, i think people are too focused on the Yamnaya Culture. We already see EHG in Areni individuals, Z2103 could have come as early as 4000 BC in south caucasus, totally unrelated with later IE migrations, or maybe related with a very early migration, but those guys did not came back into the steppe to found Yamnaya. If you insist in putting Yamnaya and PIE in the topic ( wich i dont say or even think you are doing ) it might sound like an ethnogenesis pride, such as some armenian and kurdish guys have in those topics and communities.

Let's focus on ancient dna, all questions will be answered at some point.

halfalp
17-05-19, 08:37
This old study has to be analyzed and observed in different angles. Especially considering 1) their datas 2) their conclusion 3) the relation of both 1 and 2 points with the ancient dna that we have since 2012.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3286660/

Pip
17-05-19, 12:19
IE Anatolian speakers were not steppe related (autosomal), hence my best bet go for PF7562, which is an early bronze age split from its branch sister L23. The best place of this split is northern Caucasus.
My estimates suggest this is likely, although perhaps a little sooner.

Pip
17-05-19, 12:27
Besides, Steppe Maykop had a much more "eastern" and quite probably Central Asian Steppe autosomal makeup, very distinct from that of "truly" Caucasian Maykop. I wouldn't analyze these two cultures and peoples together in terms of genetic origins and possibly even ethnic and linguistic identity.
I was just responding to the claim that there was no Maykop Z2103; but, yes, it's not especially relevant.
Having said that, specifically the Z2103 sample in Steppe Maykop is one that has the typical "true" Maykop admixture, with a substantial Anatolian component.

Pip
17-05-19, 12:44
But as far as y-dna R1, i think it's almost clear that it came from eastern europe, even Narasimhan apparently had after few C14 fails to redate the Z2103 individual from Hajji Firuz something like MLBA. Everything tends to show that prior 3000 BC, there was no R1 south of caucasus, but that after, something triggered their movement.
It's not a matter of whether R1 per se came from Eastern Europe; I'm not disputing that. The question is where the most recent common ancestor of Z2103 lived.
Nothing can ever possibly show that there was no R1 South of the Caucasus, unless we have access to the remains of every man who has lived there prior to 3,000 BC. And you talk as if there were a block of R1 people who were all aware of each other's yDNA and made a point of sticking together as a unit without exception for tens of thousands of years, each of them avoiding ever straying South of the Caucasus. This is simplistic and misleading.

Pip
17-05-19, 13:25
How would R1b-Z2103 ( if it was born and came from South Caucasus ) and with a majority of Farmer mtdna, be that high in EHG ancestry?
At the birth of Z2103 (which incidentally happened over a period of probably several hundred years), we have no way of knowing what mtDNA it had, nor whether it was high in EHG ancestry, nor whether some of its bearers had very different mtDNA and aDNA to each other. You are trying to compare yDNA in the 5th millennium BC with aDNA and mtDNA from 50 or more generations later when there is no evidential or logical basis for doing so.


Q1a2 is absent of South Caucasus modern dna, why would STR variance change anything about the deduction of its origin not there.
Because the greatest STR variances between Q1a2 samples of specific phylogeny consistently occur further North.


While R1b-Z2103 had multiple founder effects in south caucasus for 5000 years, you are talking of denial, but this is a huge one. Absolutely all scientists using your hypothesis and statistics using modern dna were wrong for now.
A very sweeping statement, and I very much doubt there is anything to prove that anyone was wrong. Besides which, much of the precise phylogeny and the volume of data has emerged only recently, so the estimates are going to be subject to frequent revision.


Your same hypothesis made scientists be clear in 2011 that R1b-M269 came in europe with LBK. Your same hypothesis made scientists be clear that R1a-M417 had a diversification origin in Iran.
You are too black and white. Nothing was 'clear' in 2011, neither is it that 'clear' now. And the whole concept that M269 'came' to Europe at any one specific time is misleading - there were probably bearers of various branches of M269 moving in and out of various parts of Europe at various times (some of which died out, some of which didn't).


Why trying to get a response only by modern dna conclusions?
Because, as things stand, modern data is far more reliable, far more precise, far more detailed, in far greater quantity, more universally published and its sampling is much more random.

Pip
17-05-19, 13:48
You are starting to show impatience over the topic like " you want it to be south of the caucasus ". Bias can be hard times. If nobody would have ramble on the Hajji Firuz R1b-Z2103, we probably would have accepted the idea that this was the " ancestor of yamnaya guys " like Markod said impulsively and with joy at the preprint of the paper, wich also show'd his bias.
I'm not bothered where it is. I'm English, have no idea whether I am Z2103, have no reason to have any preference which side of the mountains it originated, and my estimates on this have changed over the past couple of years as new data has been published.


I dont believe to probabilities and statistics in ancient and modern dna. My point is, i believe prehistoric steppe Z2103 are older than any living Z2103. All modern southern caucasus Z2103 were born in south caucasus post-chalcolithic, mountaneous region are creating lots of founder effects very fast, as for very populous regions. India show the same kind of Z93 founder effects as Z2103 south of caucasus. Being an outlier depend on the ancestral component % according to the whole population analyzed, so it's not relevant with lineage.
I do believe in data and in statistics, and have no fixed faiths or beliefs that certain haplogroups originated anywhere in particular.


Let's focus on ancient dna, all questions will be answered at some point.
Why wait? I could be dead by the time they find anything useful, become bothered to analyse it, get the funding to analyse it, see whether they can read it and decide whether to publish it. It it passes through all these hoops and then does provide useful information, it will then probably be written off as a 'blip', 'noise', an 'outlier' or of insufficient quality to be worthy of consideration.

halfalp
17-05-19, 18:29
I'm not bothered where it is. I'm English, have no idea whether I am Z2103, have no reason to have any preference which side of the mountains it originated, and my estimates on this have changed over the past couple of years as new data has been published.


I do believe in data and in statistics, and have no fixed faiths or beliefs that certain haplogroups originated anywhere in particular.


Why wait? I could be dead by the time they find anything useful, become bothered to analyse it, get the funding to analyse it, see whether they can read it and decide whether to publish it. It it passes through all these hoops and then does provide useful information, it will then probably be written off as a 'blip', 'noise', an 'outlier' or of insufficient quality to be worthy of consideration.

Do what you do and what you want. I will just not follow the same path. The real importance is to know why we are, individualy doing it for.

Pip
18-05-19, 18:17
Do what you do and what you want. I will just not follow the same path. The real importance is to know why we are, individualy doing it for.

For me, curiosity.

halfalp, you might well be right, although the specific hypothesis that living Z2103 originated in the Steppe is untestable through ancient DNA, and is thus what Popper defined as pseudoscience - a matter of instinct, much like the tenets of psychoanalysis, a belief in God or interpretations of literature. The only objective methods of testing this are currently using variance analysis on modern DNA, and they yield a slightly different alternative 'most likely' answer. But all might change as the data accumulates.

There are four basal branches of living Z2103 - three of these branches are estimated to be wholly of a South of the Caucasus coalescence (PF331, Y13369 and Y4364) and the fourth is estimated to have a Caucasus branch (CTS8966) and a Steppe/Eastern European branch (Z2109). The only steppic branch appears to be a sub-branch of a basal branch, and this sub-branch Z2109 shows little or no sign of migrating back South of the Caucasus at any subsequent point.

The Steppe looks from the data to be at best a tertiary development of Z2103, with any migrational separation across the Caucasus occurring at an early stage (whichever way it occurred), long before the late Bronze Age steppic migrations into Southern Asia. Accordingly, I would suggest these late Bronze Age steppic migrations were of people that were predominantly R1a-Z93 (rather than R1b-Z2103), albeit with probably some Yamnayan-descendant admixture.

halfalp
18-05-19, 18:47
I have hard time to imagine that the 400-200 BC Scythians Z2103 samples that we have are not somehow related with any modern branches of Z2103.

According to Maciamo then itself according to FTDNA:


- L584 (including L943) : found mostly in the South Caucasus, Turkey, Iraq and Iran, but also in Poland, Germany and Austria.

- L277.1 : found in Russia, Central Asia, Bulgaria, India and the Middle East (Iraq, Lebanon).

- CTS7822 (including CTS9219) : found in Russia (including Chuvashia), Ukraine, Poland, Hungary, the Balkans, Armenia, Italy, Spain, Britain and Ireland.

- CTS7763 : found in Turkey, Bulgaria and Italy.

- Z2109 : found in Germany and Scotland.

All South Caucasus Z2103 branches are also found in Europe. Bashkirs and Udmurts have probably more Z2103 % than ethnic Armenians but i cannot found anything on this and the potential branches they are in.

Pip
18-05-19, 23:43
I have hard time to imagine that the 400-200 BC Scythians Z2103 samples that we have are not somehow related with any modern branches of Z2103.
Do we have further information on their SNPs? If we do, we can surely find out; if we do not, we will not know one way or the other. Personally, I have no difficulty imagining that the lineage of an ancient person might have died out; it happened all the time.


According to Maciamo then itself according to FTDNA:

- L584 (including L943) : found mostly in the South Caucasus, Turkey, Iraq and Iran, but also in Poland, Germany and Austria.

- L277.1 : found in Russia, Central Asia, Bulgaria, India and the Middle East (Iraq, Lebanon).

- CTS7822 (including CTS9219) : found in Russia (including Chuvashia), Ukraine, Poland, Hungary, the Balkans, Armenia, Italy, Spain, Britain and Ireland.

- CTS7763 : found in Turkey, Bulgaria and Italy.

- Z2109 : found in Germany and Scotland.

All South Caucasus Z2103 branches are also found in Europe. Bashkirs and Udmurts have probably more Z2103 % than ethnic Armenians but i cannot found anything on this and the potential branches they are in.

Yes, as the current data stands, the South/Caucasus Z2103 samples are more similar to European samples (both in terms of SNPs and STRs) than they are to each other, suggesting that the European samples descended more recently from South/Caucasus people than when the South/Caucasus samples branched away from each other. If you analyse all of the ftDNA samples in detail and according to their precise SNPs, I think you will find the same. (L584, L277 and CTS7763 are each subclades of these branches.)

The limited information that I have on Bashkir samples places them on the European basal branch, which my estimates suggest most likely underwent its early development on the Central Steppe (the European branch being the Z2019 sub-subclade of Z2103, from which most European Z2103 branches, including CTS7822, descend).

halfalp
19-05-19, 09:59
Do we have further information on their SNPs? If we do, we can surely find out; if we do not, we will not know one way or the other. Personally, I have no difficulty imagining that the lineage of an ancient person might have died out; it happened all the time.



Yes, as the current data stands, the South/Caucasus Z2103 samples are more similar to European samples (both in terms of SNPs and STRs) than they are to each other, suggesting that the European samples descended more recently from South/Caucasus people than when the South/Caucasus samples branched away from each other. If you analyse all of the ftDNA samples in detail and according to their precise SNPs, I think you will find the same. (L584, L277 and CTS7763 are each subclades of these branches.)

The limited information that I have on Bashkir samples places them on the European basal branch, which my estimates suggest most likely underwent its early development on the Central Steppe (the European branch being the Z2019 sub-subclade of Z2103, from which most European Z2103 branches, including CTS7822, descend).

But let me just summarize your hypothesis here:

You believe the R1b-Z2103* from prehistoric steppe is a dead lineage, somehow related with the " Z2103 european basal branche " that separated from " multiple South Caucasus Z2103 basal branches " before Yamnaya Culture?

If so, what is your hypothesis of how did it came into Steppe and when?

Let's have an alternative hypothesis. In the Caucasus paper there is 2 interesting North Caucasus samples. PG2001 and PG2004 both are 4300-4000 BC old and are probably R1b-V1636*.

I'm pretty sure R1b-V1636 is nowadays only found in South Caucasus, and STR analysis would show just as R1b-Z2103 that it was probably born in modern Armenia or somwhere South of Caucasus.

Then we have 3500-3000 BC samples all over Eastern Europe that shows Z2103*.

Now let's cut the apple in two. What if, V1636 and Z2103 are originating in North Caucasus but somehow had ties with South Caucasus ( Areni ) and exchanged genes already 4000BC. Z2103 and V1636 does appear in South Caucasus wich indirectly confirms those exchanges, but it only appears post-3000 BC mainly Late Bronze Age.

If we go back the Phylogenetic Tree of R1b, V1636 is the brother of P297 and Z2103 is the grand-son of M269. It's hard to imagine there was a mutual origin both South of the Caucasus and in broad Eastern Europe since at least Paleolithic. Z2013 from Yamnaya have probably nothing to do with modern south caucasus branches and probably dont come from South Caucasus. However, it's possible that around 4000 BC some North Caucasus Z2103 and V1636 expanded South of the Caucasus before the founder effect creating the Steppe Z2103 ( probably with the cultural impulse of Maykop expansion ) came to birth.

Pip
19-05-19, 19:05
But let me just summarize your hypothesis here:
You believe the R1b-Z2103* from prehistoric steppe is a dead lineage, somehow related with the " Z2103 european basal branche " that separated from " multiple South Caucasus Z2103 basal branches " before Yamnaya Culture?
If so, what is your hypothesis of how did it came into Steppe and when?

I would guess that the surviving Z2103* lineage might well have spent part (or perhaps even much) of its lifespan in the Steppe, but there is no data to confirm this for sure, and I have no hypothesis concerning how or when it did so.

My estimate for the steppic Z2109 branch calculated as 4,200 BC (a little earlier than yfull's estimate), although I am not confident of the precision of this estimate (just that its separation from Southern branches appears to precede Yamnaya).

My guess is that it connects in some way to the steppic-Anatolian hybrid people (Suvorovo) that turned up in 5th millennium BC Bulgaria.


Let's have an alternative hypothesis. In the Caucasus paper there is 2 interesting North Caucasus samples. PG2001 and PG2004 both are 4300-4000 BC old and are probably R1b-V1636*.
I'm pretty sure R1b-V1636 is nowadays only found in South Caucasus, and STR analysis would show just as R1b-Z2103 that it was probably born in modern Armenia or somwhere South of Caucasus.
Then we have 3500-3000 BC samples all over Eastern Europe that shows Z2103*.

There are two assumptions here of which I would appreciate clarification - (i) how do we arrive at 'probably R1b-V1636'?, and (ii) what exactly is meant by Z2103*? (A definition of Z2103* from several years ago might be different to Z2103* now, due to recent confirmation of phylogeny).


Now let's cut the apple in two. What if, V1636 and Z2103 are originating in North Caucasus but somehow had ties with South Caucasus ( Areni ) and exchanged genes already 4000BC. Z2103 and V1636 does appear in South Caucasus wich indirectly confirms those exchanges, but it only appears post-3000 BC mainly Late Bronze Age.
If we go back the Phylogenetic Tree of R1b, V1636 is the brother of P297 and Z2103 is the grand-son of M269. It's hard to imagine there was a mutual origin both South of the Caucasus and in broad Eastern Europe since at least Paleolithic. Z2013 from Yamnaya have probably nothing to do with modern south caucasus branches and probably dont come from South Caucasus. <strong>However, it's possible that around 4000 BC some North Caucasus Z2103 and V1636 expanded South of the Caucasus before the founder effect creating the Steppe Z2103 ( probably with the cultural impulse of Maykop expansion ) came to birth.

Yes, indeed very possible. Still not quite the most likely in my view. I see early Z2103 as possibly moving around a focal point in the Western Caucasus, perhaps even seasonally; I also see many successful ancient lineages as cross-cultural, which can catalyse development, although this is only speculation in this case. I am not confident enough to be any more definite in this respect.

My main point of curiosity was to try to establish whether South of Caucasus Z2103 was substantially connected to later Bronze Age/Aryan migrations, and I have come to the tentative conclusion that probably wasn't and that it was most likely R1-Z93 that brought in the bulk of Steppe DNA at that point.

halfalp
19-05-19, 20:27
There are two assumptions here of which I would appreciate clarification - (i) how do we arrive at 'probably R1b-V1636'?, and (ii) what exactly is meant by Z2103*? (A definition of Z2103* from several years ago might be different to Z2103* now, due to recent confirmation of phylogeny).

1) Smal from Anthrogenica made calls for this paper several times ago here https://docs.google.com/spreadsheets/d/1wkLjTA856nW6On8Q10U-WrhyTBWlwCVdK6AMQFCfIaw/edit#gid=202340943. Im not sure if the calls are from him or the authors, but i'm pretty sure they are not wrong.

Also about R1b-Z2103*, pretty sure all Z2103 from Yamnaya and prehistoric Steppe are labeled Z2103*


Yes, indeed very possible. Still not quite the most likely in my view. I see early Z2103 as possibly moving around a focal point in the Western Caucasus, perhaps even seasonally; I also see many successful ancient lineages as cross-cultural, which can catalyse development, although this is only speculation in this case. I am not confident enough to be any more definite in this respect.

My main point of curiosity was to try to establish whether South of Caucasus Z2103 was substantially connected to later Bronze Age/Aryan migrations, and I have come to the tentative conclusion that probably wasn't and that it was most likely R1-Z93 that brought in the bulk of Steppe DNA at that point.

Now it's the point where i want to come to. Do we speak about R1b-L23 and Z2103 clearly been born from M269 in South Caucasus, or are we talking as i said about cross-cultural developpement? I have the conviction, wich might changes with new datas, that the Z2103 from the Steppe mainly Yamnaya are local, wich doesn't mean the Z2103 from Hajji Firuz for exemple are directly descending from Yamnaya guys. If Z2103 have a TMRCA older than 4000 BC, then it can clearly have lived both side of the Caucasus before Yamnaya expansions.

Pip
19-05-19, 22:58
1) Smal from Anthrogenica made calls for this paper several times ago here https://docs.google.com/spreadsheets/d/1wkLjTA856nW6On8Q10U-WrhyTBWlwCVdK6AMQFCfIaw/edit#gid=202340943. Im not sure if the calls are from him or the authors, but i'm pretty sure they are not wrong.
Thanks. Not sure that V1636 is highly relevant, though, as yfull estimates that it split from Z2103's ancestor more than 10,000 years beforehand.


Also about R1b-Z2103*, pretty sure all Z2103 from Yamnaya and prehistoric Steppe are labeled Z2103* I can only find that it is either Z2103 unspecified or I2a2a1b1, apart from one dead end L23 lineage negative for both Z2103 and L51.


Now it's the point where i want to come to. Do we speak about R1b-L23 and Z2103 clearly been born from M269 in South Caucasus, or are we talking as i said about cross-cultural developpement? I have the conviction, wich might changes with new datas, that the Z2103 from the Steppe mainly Yamnaya are local, wich doesn't mean the Z2103 from Hajji Firuz for exemple are directly descending from Yamnaya guys. If Z2103 have a TMRCA older than 4000 BC, then it can clearly have lived both side of the Caucasus before Yamnaya expansions.
I wouldn't say it is clear where these haplogroups were formed, merely that they were most likely in the general vicinity of the Western Caucasus as they were forming. Perhaps the two hypotheses are not mutually exclusive, but that the successful people bearing this yDNA moved between two environments, sheltering in and trading with farmers in the South during the winters and venturing North in the summers. When two cultures come together, this can confer the advantages of adaptability and an increased skillset.

I have mentioned this elsewhere, but I see curious associations between two diverse haplogroups R-L23 and G-PF3345, which each appear to have thrived (on both sides of the Pontic) while relatives around them were dying out. Perhaps further indications of collaboration?

halfalp
20-05-19, 08:53
I wouldn't say it is clear where these haplogroups were formed, merely that they were most likely in the general vicinity of the Western Caucasus as they were forming. Perhaps the two hypotheses are not mutually exclusive, but that the successful people bearing this yDNA moved between two environments, sheltering in and trading with farmers in the South during the winters and venturing North in the summers. When two cultures come together, this can confer the advantages of adaptability and an increased skillset.

I have mentioned this elsewhere, but I see curious associations between two diverse haplogroups R-L23 and G-PF3345, which each appear to have thrived (on both sides of the Pontic) while relatives around them were dying out. Perhaps further indications of collaboration?

Well its clear already 5000BC that Pontic Steppe and South Caucasus are exchanging females via the mtdna results we have cf. ( H2a1 in Khvalynsk and H2a1a in Sredny Stog and maybe y-dna such as J1* in Karelia. I dont think any R1b and G population ( together ) ever existed anywhere. I think J2b might be a better southern lineage to be related and found alongside R1b. It's clear that the R1b from Steppe neighbored a lot of CHG people, looking at their ancestry, even if all CHG from Steppe came via women, they had those women because of good relationship with the southern people.

And i wouldn't take too much reltation from ancient dna distribution with modern one. I have already made the research and the L-L595 haplogroup from Maykop is nowadays mainly an European lineage especially extreme west europe as Ireland and British Islands but also Estonia. Wich tells us this lineage since Maykop expanded a lot, probably with IE speakers somehow.

I think there was infiltration in Steppe from the south both East and West of the Caucasus mountains. Even though the question is, Where did R1b-Z2103 originated?

Then there is that argument that Z2103 branches of Armenia separated before the Steppic and Eastern European ones. That's probably true, but it doesn't tell us where it did separated, Armenia is just the modern focal point, not the point of origin.

halfalp
20-05-19, 09:00
Thanks. Not sure that V1636 is highly relevant, though, as yfull estimates that it split from Z2103's ancestor more than 10,000 years beforehand.

Its only relevant in the way that, if we would only have modern V1636 samples, our primary deduction would be it came from Anatolia, while it is now found in Khvalynsk Culture and North Caucasus ante-4000 BC while it is found in prehistoric South Caucasus in Late Kura-Araxes only 2600-2400 BC. But probably STR variance would put V1636 origin in Anatolia, while it's probably not. I like to think lineages dont really spread with inconsistant patterns, and R1b in Europe since Paleolithic seem a coherent pattern from now. Then there is people who believe that there was some R1b in Europe, while there was some V88 and M269 in Northern Mesopotamia at the same time... i dont believe such patterns, especially when they are incoherent in relation with the phylogenetic tree and the prehistoric datas.

Pip
20-05-19, 10:41
Its only relevant in the way that, if we would only have modern V1636 samples, our primary deduction would be it came from Anatolia, while it is now found in Khvalynsk Culture and North Caucasus ante-4000 BC while it is found in prehistoric South Caucasus in Late Kura-Araxes only 2600-2400 BC. But probably STR variance would put V1636 origin in Anatolia, while it's probably not. I like to think lineages dont really spread with inconsistant patterns, and R1b in Europe since Paleolithic seem a coherent pattern from now. Then there is people who believe that there was some R1b in Europe, while there was some V88 and M269 in Northern Mesopotamia at the same time... i dont believe such patterns, especially when they are incoherent in relation with the phylogenetic tree and the prehistoric datas.
Yes, but the quantity of V1636 data is too small from which to draw reliable conclusions. If something is not found, this does not mean it did not exist. And in human behaviour/social sciences, it is unrealistic to expect wholly coherent patterns.

Pip
20-05-19, 12:08
Well its clear already 5000BC that Pontic Steppe and South Caucasus are exchanging females via the mtdna results we have cf. ( H2a1 in Khvalynsk and H2a1a in Sredny Stog and maybe y-dna such as J1* in Karelia.
Yes, although this isn't really exchange, but signs of one-way traffic. If there was admixture between North and South of Caucasus populations, the two populations must have met, so I don't see why there would not also have been some gene flows in the opposite direction.


I dont think any R1b and G population ( together ) ever existed anywhere.
I think it likely, with G-PF3345 specifically - in the Eastern Balkans/Carpathians (and spilling over into Central/Western Europe) with L51, and in the Western Caucasus with Z2103. It might help explain why L23 and PF3345 are the only subclades of each haplogroup that thrived, and PF3345 provided what I expect living L23 acquired the majority of its Anatolian component from.


And i wouldn't take too much reltation from ancient dna distribution with modern one. I have already made the research and the L-L595 haplogroup from Maykop is nowadays mainly an European lineage especially extreme west europe as Ireland and British Islands but also Estonia. Wich tells us this lineage since Maykop expanded a lot, probably with IE speakers somehow.
Exactly, it seems to mirror R1b-L23. Probably not a coincidence, in my opinion.


I think there was infiltration in Steppe from the south both East and West of the Caucasus mountains. Even though the question is, Where did R1b-Z2103 originated?

Then there is that argument that Z2103 branches of Armenia separated before the Steppic and Eastern European ones. That's probably true, but it doesn't tell us where it did separated, Armenia is just the modern focal point, not the point of origin.

I think we are close enough to the answer - that its formative period came to a conclusion most likely somewhere near to the Caucasus. This fits with yDNA, mtDNA and aDNA evidence.

To expect much more than this is unrealistic. The formation of Z2103 as we know it represents several hundred years of lineage through a series of single individuals who underwent 10 different SNP mutations. It is impossible for us to find the remnants of each of these individuals, and track exactly where they were all conceived and where they travelled over their lifetimes.

halfalp
20-05-19, 17:30
To expect much more than this is unrealistic. The formation of Z2103 as we know it represents several hundred years of lineage through a series of single individuals who underwent 10 different SNP mutations. It is impossible for us to find the remnants of each of these individuals, and track exactly where they were all conceived and where they travelled over their lifetimes.

I think my point is that i dont believe M269 was in Balkans or Steppe, came to Anatolia or South of the Caucasus and back migrate to Steppe again. If L23 and Z2103 are born in Armenia, then L51 probably also, but a thing to keep in mind, if Z2103 was 4000BC in Armenia, its genetic signature would be 50/50 Anatolian_Chl and Iran_Neolithic. Both signature dont exist in the Z2103 Steppe samples, only CHG that some guys try to link with Iran_Neo but the thing is many calculators shows how this ancestry is closer to Kotias than Ganj Dareh or Hotu for exemple. P297 is virtually the father of M269 while M73 is the virtual brother of it. So P297 in Baltic, M73 East of the Urals, it make sense for M269 to be from the Steppe, L23 and Z2103 probably had TMRCA at some point very close to Caucasus, while L51 was somewhere else.

Pip
20-05-19, 22:38
I think my point is that i dont believe M269 was in Balkans or Steppe, came to Anatolia or South of the Caucasus and back migrate to Steppe again.

What grounds do you have for not believing that? yfull estimates M269's formation period to be 6,900 years long. Even before it started branching, there were about 230 generations of men bearing an M269-equivalent SNP. Why would they all have stayed in exactly the same spot for the whole of their lifespans? M269 would have been formative over a zone, almost certainly a wide one.
Some were migrating backwards and forwards all the time. Suvorovo seem to have moved from the Central Steppe to the Balkans and back, and R1a-Z283 appears to have migrated from the Steppe to Germany and back again. It is not unheard of.


if Z2103 was 4000BC in Armenia, its genetic signature would be 50/50 Anatolian_Chl and Iran_Neolithic. Both signature dont exist in the Z2103 Steppe samples
Armenia was probably inhabited by various populations with different signatures; I can't see that anyone present there would automatically have been 50:50 Anatolian:CHG.
But in any case, within the next 1,000 years by the time we get to the Z2103 Steppe samples, the original Armenian autosomal mix would have been diluted by 30 or more admixtures with females, potentially creating various signatures, each unrecognisable from their start point. That is why we have early Z2103 samples with a variety of autosomal mixes - some heavily EHG, some heavily CHG, one heavily Anatolian, one heavily WHG.

P297 is virtually the father of M269 while M73 is the virtual brother of it. So P297 in Baltic, M73 East of the Urals, it make sense for M269 to be from the Steppe, L23 and Z2103 probably had TMRCA at some point very close to Caucasus, while L51 was somewhere else.
Ancient P297 and M73 in the Baltic are red herrings - the lineages of all of these samples are extinct, suggesting that the Baltic was not the place most conducive to the emergence of successful P297 lineages.
I would suggest the main zone within which Z2103 formed was most likely the Eastern Pontic between Laz and Azov. However, in my view, it would not be out of the question for Z2103 to have also been within the group that moved in and destroyed the Chalcolithic settlements in the Balkans and Carpathians before migrating back to the East again.

halfalp
21-05-19, 09:52
What grounds do you have for not believing that? yfull estimates M269's formation period to be 6,900 years long. Even before it started branching, there were about 230 generations of men bearing an M269-equivalent SNP. Why would they all have stayed in exactly the same spot for the whole of their lifespans? M269 would have been formative over a zone, almost certainly a wide one.
Some were migrating backwards and forwards all the time. Suvorovo seem to have moved from the Central Steppe to the Balkans and back, and R1a-Z283 appears to have migrated from the Steppe to Germany and back again. It is not unheard of.


Armenia was probably inhabited by various populations with different signatures; I can't see that anyone present there would automatically have been 50:50 Anatolian:CHG.
But in any case, within the next 1,000 years by the time we get to the Z2103 Steppe samples, the original Armenian autosomal mix would have been diluted by 30 or more admixtures with females, potentially creating various signatures, each unrecognisable from their start point. That is why we have early Z2103 samples with a variety of autosomal mixes - some heavily EHG, some heavily CHG, one heavily Anatolian, one heavily WHG.

Ancient P297 and M73 in the Baltic are red herrings - the lineages of all of these samples are extinct, suggesting that the Baltic was not the place most conducive to the emergence of successful P297 lineages.
I would suggest the main zone within which Z2103 formed was most likely the Eastern Pontic between Laz and Azov. However, in my view, it would not be out of the question for Z2103 to have also been within the group that moved in and destroyed the Chalcolithic settlements in the Balkans and Carpathians before migrating back to the East again.

You keep comparing modern branches and considering prehistoric ones as " extinct ". People did the same with Villabruna or Iron_Gates / Baltic HG's.

Let me just resume on what your hypothesis is holding taking ancient dna in account. Because there is no R1b ante-3000 BC in Anatolia, South Caucasus or broad Middle-East. Apparently according to almost everyone on Eupedia, R1b-V88, R1b-M269, R1b-L23, R1b-Z2103, R1b-L51 were all Anatolian-broad Middle-East lineages but apparently we cannot found any of them. Apparently the Europeans samples are Dead Lineages so Irrelevant and apparently the lineages that survived cannot be found because they were ultra minor lineages in Anatolia-Broad Middle-East until they reach Europe and then expanded like crazy.

The diversity of prehistoric Armenia probably was not diverse as you would want it to be. Yamnaya samples have Z2103 and mostly souther mtdna, were the 60% EHG came from? Once again its shadow ultra minor lineage that would have expanded like crazy. I dont buy it.

About your point about Anatolian and WHG heavy Z2103, it's very symptomatic of transitional peoples. Meaning the Anatolian ancestry of Northern Caucasus samples are totally irrelevant for the history of Z2103 and more relevant in the history of Maykop People lacking Z2103. Most of none North Caucasus Z2103 are contemporary with it and do not show Anatolian ancestry, meaning this ancestry never expanded over the Manych River. As the same with the extra WHG ancestry West. M73 was not found in the Baltic as far as i remember but in Botai, wich totally match the geography of modern distribution east of the Urals. We dont have much data of modern P297 but its probably born not far of the Baltic and if not Baltic probably east of it.

If there is enough room to say that everything about the ancient dna we have is irrelevant about prehistoric history ( what? ), there is not enough room to talk about all this. We can just write the history with modern dna. But then why would ancient dna exist and paper about it exist if modern dna alone was relevant? Thats a good paradoxe that we should do probabilities of ancient dna with modern dna.

Pip
22-05-19, 09:33
The difference is I am assuming lineages are dead when they do not appear amongst tens of thousands of modern samples; you are assuming that lineages did not exist when they do not appear amongst only tens of ancient samples.
Ancient DNA is not irrelevant though - it can be very informative when it is potentially a living lineage and can be traced into modern populations. Yamnayan Z2103 is a good example of this, but I don't think ancient DNA tells us much about the formation of basal M269 clades, as the samples aren't really there at early enough dates. Consequently, we can't really say what the autosomal mixes were in their earliest populations, or what might have been admixed in by just one or two generations of females from other populations.
Ancient DNA tells us more about the formation of R1a-M417, as a couple of samples fall close to formation dates, have consistent autosomal readings and can be traced through well into living populations.

halfalp
23-05-19, 09:35
The difference is I am assuming lineages are dead when they do not appear amongst tens of thousands of modern samples; you are assuming that lineages did not exist when they do not appear amongst only tens of ancient samples.
Ancient DNA is not irrelevant though - it can be very informative when it is potentially a living lineage and can be traced into modern populations. Yamnayan Z2103 is a good example of this, but I don't think ancient DNA tells us much about the formation of basal M269 clades, as the samples aren't really there at early enough dates. Consequently, we can't really say what the autosomal mixes were in their earliest populations, or what might have been admixed in by just one or two generations of females from other populations.
Ancient DNA tells us more about the formation of R1a-M417, as a couple of samples fall close to formation dates, have consistent autosomal readings and can be traced through well into living populations.

Just as an exemple, do we know if the Cogotas R1b-DF27 falls into modern DF27 branches or is it a dead lineage branch? If it's a dead one, what does it have different than Z2103 samples from the Steppe?

Yes you maybe right about R1a-M417, but before we had ancient dna of them, people used to force the origin of it in Kurdistan with the same technology you are making Z2103 coming from Armenia, so why are they wrong because of Ancient DNA and you dont because of Modern DNA? Only because they are close to formation date?

Pip
23-05-19, 15:36
Just as an exemple, do we know if the Cogotas R1b-DF27 falls into modern DF27 branches or is it a dead lineage branch? If it's a dead one, what does it have different than Z2103 samples from the Steppe?
Yes you maybe right about R1a-M417, but before we had ancient dna of them, people used to force the origin of it in Kurdistan with the same technology you are making Z2103 coming from Armenia, so why are they wrong because of Ancient DNA and you dont because of Modern DNA? Only because they are close to formation date?
I'll look at Cogotas if I can find data on it. If it's like other BA Iberian I've seen, it will be a mix of Northern Bell Beaker, El Portalon and Iberian Neolithic.
Not sure why Kurdistan for M417 - perhaps they had little data or applied unsophisticated methodology or were looking at R1a as a whole. My calculations for MRCA gave SE Poland/NW Ukraine, but still based on fairly little data. Still, it ties up roughly with ancient DNA, which gives an early NE Ukraine sample with a semi-Balkan like autosomal profile.

nornosh
23-05-19, 16:55
Question: percentage of R1b-Z2103 in Western Iran, Turkmenistan, in Wikipedia map it shows 30-60% in W.Iran is it true?

Pip
24-05-19, 13:28
Question: percentage of R1b-Z2103 in Western Iran, Turkmenistan, in Wikipedia map it shows 30-60% in W.Iran is it true?
Where is this map? I doubt it. Perhaps at the lower end of this range in the districts bordering Armenia, but I don't think so as a whole.

Pip
24-05-19, 13:45
Just as an exemple, do we know if the Cogotas R1b-DF27 falls into modern DF27 branches or is it a dead lineage branch? If it's a dead one, what does it have different than Z2103 samples from the Steppe?
Cogotas looks like a dead lineage, although its nearest modern relatives are also Iberian (estimated split point - late 3rd millennium BC, per yfull). It has an autosomal profile that looks similar to other Bronze Age Iberia (a mix of Northern Bell Beaker, El Portalon and Iberian Neolithic). Is this a trick question?

nornosh
24-05-19, 17:16
Where is this map? I doubt it. Perhaps at the lower end of this range in the districts bordering Armenia, but I don't think so as a whole.

the map is in R1b article. in Turkmenistan it shows very high figures too yet some say its mistaken with hg Q but I say how come they mistake it for hg Q which is so different you can't just mistake different haplogroups in tests so it must be R1b. I have seen some Turkmen pictures they do resemble Eurasian populations more.
by the the way in Kabul, Afghanistan in general there are big populations of Iranic people do they share similar mixture with Iranians( R1b, J2,...) or not? I think they(we) could be R1b-heavy too plz provide me some links anyone.

halfalp
24-05-19, 19:30
Cogotas looks like a dead lineage, although its nearest modern relatives are also Iberian (estimated split point - late 3rd millennium BC, per yfull). It has an autosomal profile that looks similar to other Bronze Age Iberia (a mix of Northern Bell Beaker, El Portalon and Iberian Neolithic). Is this a trick question?

Not necessarily a trick question, but a proof that a dead lineage from a modern branch can be prehistorically in his modern distribution. All R1b from Ukraine are mostly Z2103, even tho their branch is not the one found in Yamnaya, its related just the TMRCA is younger than Yamnaya just as the Cogotas one. Probably there will never be ancient dna from some Armenian R1b of Chalcolithic that already have the modern living branches found there.

halfalp
24-05-19, 19:32
the map is in R1b article. in Turkmenistan it shows very high figures too yet some say its mistaken with hg Q but I say how come they mistake it for hg Q which is so different you can't just mistake different haplogroups in tests so it must be R1b. I have seen some Turkmen pictures they do resemble Eurasian populations more.
by the the way in Kabul, Afghanistan in general there are big populations of Iranic people do they share similar mixture with Iranians( R1b, J2,...) or not? I think they(we) could be R1b-heavy too plz provide me some links anyone.

If i remember from a few years, the Turkmenistan R1b falls mostly in the eastern R1b-M73 branch not Z2103.

Pip
24-05-19, 22:21
the map is in R1b article. in Turkmenistan it shows very high figures too yet some say its mistaken with hg Q but I say how come they mistake it for hg Q which is so different you can't just mistake different haplogroups in tests so it must be R1b. I have seen some Turkmen pictures they do resemble Eurasian populations more.
by the the way in Kabul, Afghanistan in general there are big populations of Iranic people do they share similar mixture with Iranians( R1b, J2,...) or not? I think they(we) could be R1b-heavy too plz provide me some links anyone.
The only likely Z2103 I can find in Turkmenistan is of the Steppic branch, rather than the South Caucasus/Iranian branches. It is very similar to samples from Ukraine and Belarus, and my guess is that it is of relatively recent Russian paternal ancestry.

Pip
24-05-19, 23:34
Not necessarily a trick question, but a proof that a dead lineage from a modern branch can be prehistorically in his modern distribution. All R1b from Ukraine are mostly Z2103, even tho their branch is not the one found in Yamnaya, its related just the TMRCA is younger than Yamnaya just as the Cogotas one. Probably there will never be ancient dna from some Armenian R1b of Chalcolithic that already have the modern living branches found there.
Of course, it is possible that a dead lineage can be found in a similar location to its most recent modern relatives - all other things being equal, this is the most likely location for it to be found. But when a sample of a dead lineage is found in a different location to its most recent modern relatives, this does not tell us much, particularly when the coalescence points of each of its most recent modern relatives are distributed over a very wide area.

An interesting ancient Z2103 sample is I5884 (2,800 BC) from Ukraine, which has some inconsistent-looking SNP readings suggestive of one of its South Caucasus branches - the one that I estimate was most recently-related to the Steppe branch. It is what would usually be called an 'outlier', but most people cannot bring themselves to call it this, as they are so wedded to the idea of Z2103 being synonymous with the Steppe and not an outlier to it. It has virtually no CHG, a huge amount of WHG, plenty of EHG and Anatolian, and appears to bear little autosomal similarity to Yamnayan Z2103 samples. How would you suggest it might have got there?

halfalp
25-05-19, 11:45
Your individual genetic signature looks way more like what an original Bell Beaker should be than a prehistoric Armenian, dont know where you want to go here. It then means that the South Caucasus branches came from Steppe following your own reasoning, just that they were slightly different than the general Yamnaya signature. Wich funny enough, gives even way more credit to the Balkans road hypothesis for Anatolians and Armenians comparing it to the Caucasus road hypothesis.

Pip
25-05-19, 14:39
Your individual genetic signature looks way more like what an original Bell Beaker should be than a prehistoric Armenian, dont know where you want to go here. It then means that the South Caucasus branches came from Steppe following your own reasoning, just that they were slightly different than the general Yamnaya signature. Wich funny enough, gives even way more credit to the Balkans road hypothesis for Anatolians and Armenians comparing it to the Caucasus road hypothesis.

I'm not going anywhere in particular, and am not tied to any particular viewpoint. As I say, I'm just curious.

This is only one sample, so we cannot conclude too much from it - but its signature must have come from somewhere, and it does not seem to fit at all with Yamnaya. (If it acquired its EHG from any variety of Yamnaya, it would have far more CHG in it.)

I've had a hard time simulating a close fit to any combination of ancient populations, but coming at it laterally I've just found a very close fit indeed - Eastern Baltic Neolithic + Balkan Chalcolithic + Globular Amphora. This would suggest that it probably was an outsider to its Dnieper location, with all three best-fit ancestral components coming from a similar longitude to its West. My guess is that it was brought from the Dniester with GAC not long beforehand (i.e. post-3,000 BC). The question is - from where would it have acquired its Z2103? As GAC looks to be principally an I2a population, the Eastern Baltic and the Eastern Balkans look the only likely candidates.

halfalp
25-05-19, 19:22
I'm not going anywhere in particular, and am not tied to any particular viewpoint. As I say, I'm just curious.

This is only one sample, so we cannot conclude too much from it - but its signature must have come from somewhere, and it does not seem to fit at all with Yamnaya. (If it acquired its EHG from any variety of Yamnaya, it would have far more CHG in it.)

I've had a hard time simulating a close fit to any combination of ancient populations, but coming at it laterally I've just found a very close fit indeed - Eastern Baltic Neolithic + Balkan Chalcolithic + Globular Amphora. This would suggest that it probably was an outsider to its Dnieper location, with all three best-fit ancestral components coming from a similar longitude to its West. My guess is that it was brought from the Dniester with GAC not long beforehand (i.e. post-3,000 BC). The question is - from where would it have acquired its Z2103? As GAC looks to be principally an I2a population, the Eastern Baltic and the Eastern Balkans look the only likely candidates.

Things to remind is that Eastern Europe is huge, very huge. Z2103 probably have a more eastern distribution while I2a2 have a western one. But that's only taking accounts samples we have, and this individual could be a more western individual with an eastern lineage.

Pip
25-05-19, 20:17
Things to remind is that Eastern Europe is huge, very huge. Z2103 probably have a more eastern distribution while I2a2 have a western one. But that's only taking accounts samples we have, and this individual could be a more western individual with an eastern lineage.
I suspect his lineage went around the Pontic and through the Balkans/Carpathians with L51, then retreated up the Dniester admixing with WHG on the Chalcolithic collapse. It is probably quite distantly related to Yamnaya.