Genetic origins of the Minoans and Mycenaeans

[Principe]

If anyone can answer this it would be much appreciated, I noticed that I9041 and I9010 Mycenaean were from Troezen and Troezen founded a colony Sybaris which is not geographically too far from my father's paternal village, I compared my dna to them and these were the results: I9041 largest segment is 6.4 cm and total of segments is 97.1 cm and I9010 largest segment 9.6 cm and total of segments is 9.6 cm, is this significant at all? Both the largest segments are on the 14th Chromsome.

Those are the results with the minimum threshold at 100 snp count

With 50 snp threshold count it changes

I9041: largest segment is now at 11.3 cm and total of segments is 909.7 cm
I9010: largest segment stays the same and the total of segments is now at 373.1 cm.

[Gravetto-Danubian]

If Kumtepe4 is real, and represents some early movement from the Balkans into Anatolia (and also into Greece, if we accept the Kumtepe4 fits for Myceneans above), then we see that EEF+CHG+EHG exists in a wide range throughout the circumpontic region, including in Yamnaya itself, and considering that EHG and CHG appeared sporadically as mixtures into Neolithic and Chalcolithic Balkans, multiple scenarios are up on the table again. Its possible that EEF was more important in the proximate IE core and a EEF+CHG+EHG mix close to the balkans that is yet unsampled, say in Suvorovo or Cernavoda, got diluted as we go East into Yamnaya territory proper. The low level of Yamnaya admixture in the Balkans post Bronze-Age would then be explained, and the Myceneans would seem to be the result of a folk migration (not conquest) from a Balkan population with already low levels of EHG.

Yes I think that has always been the most rational explanation, from one angle. But the issue of (northern) CHG introgression, and its massive intrusion west still requires explanation.

Issue for this hypothesis is the extreme sex-biased distribution of ancestry across chromosomes even in Bronze Age Balkan samples, with EEF biased towards females and EHG+CHG ancestry extremely biased towards males. Both EEF and CHG on the other hand are likewise biased towards females in those Eastern yamnaya samples we all know and love. If some Balkan input was more prominent, then EEF may be expected to be biased towards males even in Eastern yamnaya.

A further layer of complication: there are lot of non-R1 Y chromosomes in the Bronze Age Balkans. Some of these were already present in the preceding Neolithic cultures. If male-mediated sex-biased admixture took place from a population identical to Eastern Yamnaya, why is there a prevalence of I2 and other chromosomes than R1? Or, since single paternity events can change the patriline or integrate a new patriline into a social group, are we focusing too much attention on this here?

Pre-4500 BC
G2a 11; C1a - 4, H2 - 1, E1b - 1; J2a- 1 I2a2 - 3 R1b - ?V88 3 (n=24)

Post-4000 BC:
G2a -3; I2a2- 4; H2 - 1, J2a - 3; J2b2 - 1; R1a-Z93 - 1.

We're looking a complex pattern of haplogroup shifts, not monolineage. But the data set is baised toward Bulgaria at present. We'd need some propper sampling of the Dinarides and central Greece.

And I'd treat Hungary as a separate entity.

On the other hand, if we do accept Kumtepe4 as representing real movement, then we cannot associated PIE with Hittite with CHG only, as some suggest.

3) If Kumtepe4 is a fluke or contaminated, or the low quality is causing issues, we are back to the same drawing board with a proximal nucleus in Yamnaya proper, or maybe some peri-Yamnaya phenomenon in space or time to explain Anatolian. And we will wait for further sampling of Anatolia to provide for a more secure EHG signal later than Kumtepe4 itself.

Yes it could turn out 'EHG' is indeed the glue in all this.
BTW - are you using weighted or unweighted data ?

[bryan]

Given not only that Sicily was once 100% WHG and that the Mycenaeans and Dorians do not seem particularly Levant-shifted at all, it is likely Sicilians really DID get their Levantine shift from Phoenicians, Carthaginians, and the Arab conquest. There is no other explanation.

Moreover, when did Greece lose its highly "Atlantic Med" type affinity in favor of more Caucasian? That seems to be one difference between Mycenaeans and Greeks, even modern ones, in addition to that Greeks today shift noticeably toward NE Europe.

Anatolia_BA even if it's a relatively lower quality sample seems extremely similar to modern Sicilians, Greek Islanders.


# Population Percent
1 Atlantic_Mediterranean_Neolithic 44.16
2 West-Asian 24.75
3 Near_East 19.13
4 North-East-European 11.95

Single Population Sharing:

# Population (source) Distance
1 Sicilian (derived) 5.88
2 Italian-South (derived) 6.33
3 Italian-Center (derived) 7.82
4 Greek_Cretan (derived) 8.95
5 Jew_Francestrale (derived) 9.43
6 Greek_East (derived) 9.89
7 Jew_Italia (derived) 10.11
8 Greek_Center (derived) 10.64
9 Sephardim (derived) 11.03
10 Ashkenazim (derived) 11.48

# Primary Population (source) Secondary Population (source) Distance
1 86.8% Greek_Cretan (derived) + 13.2% Atlantic_Mediterranean_Neolithic (ancestral) @ 1.06
2 69.4% Cypriot (derived) + 30.6% Basque (derived) @ 3.02
3 78.4% Greek_Cretan (derived) + 21.6% Sardinian (derived) @ 3.27
4 55.1% Turk (derived) + 44.9% Sardinian (derived) @ 3.63
5 52.8% Sardinian (derived) + 47.2% Kurd (derived) @ 3.86
6 52% Sardinian (derived) + 48% Azeri (derived) @ 4.03
7 85% Greek_Cretan (derived) + 15% Gok4 (derived) @ 4.15
8 53.8% Cypriot (derived) + 46.2% Kosovar (derived) @ 4.16
9 51.6% Sardinian (derived) + 48.4% Jew-Uzbekistan (derived) @ 4.29
10 50.3% Armenian (derived) + 49.7% Sardinian (derived) @ 4.34

[ernekar]

Is there any way for me to see the Y haplogroups of these mycenean kits yet?

[lukaszM]

I'm of Greek Heritage. I ran these kits in a one to one against my own. What number of SNPs and what cenit-morgan length should I use to see how close I am to these samples?

Start from 200 SNP ansd 3 cM threshold. User Kurd recommended this to test ancient samples in one-to-one.

[lukaszM]

New upload:

I9006 Mycenaean - gedmatch M209270

[Sakattack]

Given not only that Sicily was once 100% WHG and that the Mycenaeans and Dorians do not seem particularly Levant-shifted at all, it is likely Sicilians really DID get their Levantine shift from Phoenicians, Carthaginians, and the Arab conquest. There is no other explanation.

Moreover, when did Greece lose its highly "Atlantic Med" type affinity in favor of more Caucasian? That seems to be one difference between Mycenaeans and Greeks, even modern ones, in addition to that Greeks today shift noticeably toward NE Europe.

All of the Myc samples (except Armenoi of course) score really high Caucasus, like 40%, so moderns did not extend that.

[ffoucart]

As I suggested above, the development of the Helladic period occurs from c 3500 BC (after a significant hiatus in many parts of Balkans) and a population thinning in Greece, and arrival of people bearing material culture attributes similar to Baden (Hungary) and Cernavoda-Usatavo (west Black Sea).
A second phase of impacts occurred c. 2200 BC - but this was mostly of West Asian provenance (plus some limited Vucedol influences on western Greece).

This is how the Myceneans models - no matter which exact method : roughly thirds as (i) local late Neolithic (ii) an EEF -heavy northern element containing some steppe admixture (admixture began as early as 4500 BC and continued into the late Yamnaya period - 2500 BC) (iii) West Asian/ Armenian

The date figures eerily correspond to the dating of language splits proposed by Change and Garrett: c. 3500 BC split between Indo-Aryan and Armeno-Greek and c. 2000 BC Armenian <-> Greek.

And I would guess that the most historically plausible perspective, based on recent excavations which have fine-tuned local chronologies in Greece.

Given DNA results, there is a difficulty in your chronology: the Steppe component appeared only in the last phase (after 2000 BC), as it is the big difference between Minoans and Mycenaeans.

Incursions from Anatolia in the IIId millenium BC seem connected to Iran_N/CHG, as shown in the Anatolian samples, and could be seen in the Minoans samples, meaning they are connected to the pre-IE strata.

The logical deduction is that the Steppe incursions in the Balkans before 3000 BC have little to do with Greek. Given that the Yamna derived culture found in Thrace is dated from around 3100 to 2500 BC, if we try to find a connection, it would more logical to try with the Anatolian branch of IE: Luwians, Palians and Hittites, as they are firstly attested in Central Anatolia around 2000 BC. The gap between IE populations in Thrace and in Central Anatolia would be of less than 500 years (and tens of kilometers, given than Pala was just the other side of the Bosphorus).

By the way, the split between Armenian and Greek (around 2000BC) is not backing an Eastern origin of the Steppe admixture found in Myceneans, as Armenians BA show other admixtures not found in Myceneans.

So, the logical deduction would be that the original population, which is the source for Armenian and Greek, was probably relatively high in Steppe admixture, and with limited admixture from components found in Minoans.

My guess would be Northern Balkans or nearby (Pontic Steppe?). Armenian seem also to have taken a Northern path, so the choices are limited.

But it is extremely difficult to say now, given that we probably don't have enough sampling, and probably not from the populations we need.

[Kanenas]

All of the Myc samples (except Armenoi of course) score really high Caucasus, like 40%, so moderns did not extend that.

It depends on the calculator a little. I9041 with MDLP K23b has more Caucasian than some Greeks which is the opposite of what he says.
In that calculator the Mycenean can be modeled as a mix of Sardinians and Georgians.

#
Population
Percent

Caucasian 41.3
European_Early_Farmers 36.93
Near_East 10.41
European_Hunters_Gatherers 6.83
North_African 1.4
Ancestral_Altaic 1
Melano_Polynesian 0.89
South_Central_Asian 0.63
African_Pygmy 0.61

But I think in all calculators Myceneans appear more EEF, less EHG compared to moderns and that's the main reason they are a little closer to some Italian groups.

[alexfritz]

All of the Myc samples (except Armenoi of course) score really high Caucasus, like 40%, so moderns did not extend that.

actually lower, 8-17% CHG/Iran_N it is anatolia_ba that scores CHG at >30%
http://www.nature.com/nature/journal...e23310_T1.html
i view this as akin to the sit in europe with the gradient of WHG admix in farmers from EN>MN;