Genetic Origins of Minoans and Mycenaeans

[Jovialis]

I do not have access to MTA. What modern populations is this sample closest to?

I seriously doubt this is remotely correct, but here it is:

1. Algerian_Jewish (18.13)
2. Italian_Jewish (20.13)
3. Maltese (20.31)
4. South_Italian (22.00)
5. Tunisian_Jewish (22.27)
6. West_Sicilian (22.31)
7. Greek_Islands (23.82)
8. Libyan_Jewish (23.95)

This is a very poor fit across the board.

 

[Jovialis]

Here is a 3D PCA with Dodecad Globe 13 using the samples from this study, and the Clemente et al. 2021 samples:

 

[Jovialis]

Dodecad Globe 13 for Mik15 vs available modern populations:

Distance to:	Mik15
14.79754372	Morocco_Jews
16.63085386	Sephardic_Jews
17.52413479	Sicilian_D
17.78300593	S_Italian_Sicilian_D
19.19753370	Ashkenazi_D
19.37811394	C_Italian_D
19.41410055	Ashkenazy_Jews
22.80840415	Tuscan
23.31169878	TSI30
23.33840826	Moroccan_D
24.24646985	Greek_D
25.28409184	Cypriots
25.50218226	Sardinian
25.87959235	Algerian_D
26.61235240	Canarias_1KG
26.75319233	Mozabite
26.88983637	North_Italian
27.19590594	Turkish_Cypriot_D
27.83072583	O_Italian_D
29.10995191	Andalucia_1KG
29.12918296	Moroccans
29.31107811	Murcia_1KG
30.27208780	N_Italian_D
31.58302234	Baleares_1KG
31.68168714	Portuguese_D

Here is Mik15 vs the other ancient Greek samples:

He is very close to Minoans, which we know are mostly Anatolian_N with about 15% CHG/IN. There is no way Algerian_Jews are composed of that admixture. We can't really compare these samples so much to modern populations, because the inferences made by doing so are bound to be false:

Distance to:	Mik15
4.73667605	I2937
5.72554801	I9129
5.90445594	I9010
7.59130424	Pta08
7.62219785	I9128
9.00139989	I9005
9.35317593	I0074
9.42462731	Kou01
9.80234156	I0071
11.08796194	I0073
11.33379460	I0070
11.56451901	I9130
12.19090235	I9127
12.45656453	I9131
13.06872985	I9041
13.70391915	I9006
14.52259274	I9033
15.10968564	Kou03
18.59169438	I9123
18.99046866	I2495
22.03010667	I2499
22.92758600	I2683
25.84222707	Log02
34.33169236	Log04

Autosomal-wise Mik15 looks like a Minoan, with an additional small amount of Steppe. Again, there is no way Algerian_Jews or Moroccan_Jews look like that in terms of autosomal DNA. This is why we must be careful with these tools:

 

[Archetype0ne]

Dodecad Globe 13 for Mik15 vs available modern populations:

Distance to:	Mik15
14.79754372	Morocco_Jews
16.63085386	Sephardic_Jews
17.52413479	Sicilian_D
17.78300593	S_Italian_Sicilian_D
19.19753370	Ashkenazi_D
19.37811394	C_Italian_D
19.41410055	Ashkenazy_Jews
22.80840415	Tuscan
23.31169878	TSI30
23.33840826	Moroccan_D
24.24646985	Greek_D
25.28409184	Cypriots
25.50218226	Sardinian
25.87959235	Algerian_D
26.61235240	Canarias_1KG
26.75319233	Mozabite
26.88983637	North_Italian
27.19590594	Turkish_Cypriot_D
27.83072583	O_Italian_D
29.10995191	Andalucia_1KG
29.12918296	Moroccans
29.31107811	Murcia_1KG
30.27208780	N_Italian_D
31.58302234	Baleares_1KG
31.68168714	Portuguese_D

Here is Mik15 vs the other ancient Greek samples:

He is very close to Minoans, which we know are mostly Anatolian_N with about 15% CHG/IN. There is no way Algerian_Jews are composed of that admixture. We can't really compare these samples so much to modern populations, because the inferences made by doing so are bound to be false:

Distance to:	Mik15
4.73667605	I2937
5.72554801	I9129
5.90445594	I9010
7.59130424	Pta08
7.62219785	I9128
9.00139989	I9005
9.35317593	I0074
9.42462731	Kou01
9.80234156	I0071
11.08796194	I0073
11.33379460	I0070
11.56451901	I9130
12.19090235	I9127
12.45656453	I9131
13.06872985	I9041
13.70391915	I9006
14.52259274	I9033
15.10968564	Kou03
18.59169438	I9123
18.99046866	I2495
22.03010667	I2499
22.92758600	I2683
25.84222707	Log02
34.33169236	Log04

Autosomal-wise Mik15 looks like a Minoan, with an additional small amount of Steppe. Again, there is no way Algerian_Jews or Moroccan_Jews look like that in terms of autosomal DNA. This is why we must be careful with these tools:

Fantastic visualizations Jovialis.

Yes, I think its quite normal when you have such a poor fit to modern populations. Those distances are of course gonna reflect changes in genetic makeup components from the modern populations to the ancient sample. The case still remains, that if any modern population is close (no matter how far), it seems to be some sort of Jewish continuum.

Love the rainbow by the way. Do you think the smooth gradient of ancient samples in itself, with reference to this sample, points out to how heterogeneous the Myceneans, Minoans and Ancient Greeks were?

I think myself these conversations that pop up from time to time about x ancient people was y modern population is a bit ridiculous, also knowing how these PCAs are created in theory.

Take a Portuguese Father, Russian Mother, and the kid pops up German. Might as well give him a German pass*port right?
It gets more accurate when you add more components, two + way calculators etc..., but the flaw still remains that with the same ingredients you can make different dishes that taste quite different.

 

[Jovialis]

Fantastic visualizations Jovialis.

Yes, I think its quite normal when you have such a poor fit to modern populations. Those distances are of course gonna reflect changes in genetic makeup components from the modern populations to the ancient sample. The case still remains, that if any modern population is close (no matter how far), it seems to be some sort of Jewish continuum.

Love the rainbow by the way. Do you think the smooth gradient of ancient samples in itself, with reference to this sample, points out to how heterogeneous the Myceneans, Minoans and Ancient Greeks were?

I think myself these conversations that pop up from time to time about x ancient people was y modern population is a bit ridiculous, also knowing how these PCAs are created in theory.

Take a Portuguese Father, Russian Mother, and the kid pops up German. Might as well give him a German pass*port right?
It gets more accurate when you add more components, two + way calculators etc..., but the flaw still remains that with the same ingredients you can make different dishes that taste quite different.

Indeed, an example I like to use about the possible flaws in using PCAs is the fact that South Asians, and South and Central Americans overlap. Nevertheless, we know that they are radically different in terms of their actual autosomal composition:

 

[bicicleur 2]

Indeed, an example I like to use about the possible flaws in using PCAs is the fact that South Asians, and South and Central Americans overlap. Nevertheless, we know that they are radically different in terms of their actual autosomal composition:

Because the variation is multidimensional, and PCA is just a two-dimensional projection along the 2 axes with most variation.
But it is a very easy and fast visual representation.

 

[bicicleur 2]

It's ~62% Steppe MLBA

Steppe MLBA like in Corded Ware / Sintashta ?

I think these people didn't come directly from the steppe, rather from the Carpathian Basin.

 

[Jovialis]

Because the variation is multidimensional, and PCA is just a two-dimensional projection along the 2 axes with most variation.
But it is a very easy and fast visual representation.

Indeed the z-axis on the 3D PCA demonstrates differences better.

 

[Jovialis]

The link in my signature has a good resource to demonstrate that:

https://www.eupedia.com/forum/threa...Ancient-Modern?p=615989&viewfull=1#post615989

 

[Hawk]

It's very likely those Middle Bronze Age Helladic were the actual Greek speakers. The Cycladic, Minoan and more Southern Greek were probably non-Proto-Greek Farmer + CHG mix.

I wonder what happened to G2a, during Early Neolithic they became so dominant in Europe, then again during Late Neolithic and then Bronze Age they became a relic of the past.

I was expecting E-V13 to be the dominant Neolithic Y-DNA in Balkans, but yet again G2a was all over the place.

 

[Ailchu]

Dodecad Globe 13 for Mik15 vs available modern populations:

Distance to:	Mik15
14.79754372	Morocco_Jews
16.63085386	Sephardic_Jews
17.52413479	Sicilian_D
17.78300593	S_Italian_Sicilian_D
19.19753370	Ashkenazi_D
19.37811394	C_Italian_D
19.41410055	Ashkenazy_Jews
22.80840415	Tuscan
23.31169878	TSI30
23.33840826	Moroccan_D
24.24646985	Greek_D
25.28409184	Cypriots
25.50218226	Sardinian
25.87959235	Algerian_D
26.61235240	Canarias_1KG
26.75319233	Mozabite
26.88983637	North_Italian
27.19590594	Turkish_Cypriot_D
27.83072583	O_Italian_D
29.10995191	Andalucia_1KG
29.12918296	Moroccans
29.31107811	Murcia_1KG
30.27208780	N_Italian_D
31.58302234	Baleares_1KG
31.68168714	Portuguese_D

Here is Mik15 vs the other ancient Greek samples:

He is very close to Minoans, which we know are mostly Anatolian_N with about 15% CHG/IN. There is no way Algerian_Jews are composed of that admixture. We can't really compare these samples so much to modern populations, because the inferences made by doing so are bound to be false:

Distance to:	Mik15
4.73667605	I2937
5.72554801	I9129
5.90445594	I9010
7.59130424	Pta08
7.62219785	I9128
9.00139989	I9005
9.35317593	I0074
9.42462731	Kou01
9.80234156	I0071
11.08796194	I0073
11.33379460	I0070
11.56451901	I9130
12.19090235	I9127
12.45656453	I9131
13.06872985	I9041
13.70391915	I9006
14.52259274	I9033
15.10968564	Kou03
18.59169438	I9123
18.99046866	I2495
22.03010667	I2499
22.92758600	I2683
25.84222707	Log02
34.33169236	Log04

Autosomal-wise Mik15 looks like a Minoan, with an additional small amount of Steppe. Again, there is no way Algerian_Jews or Moroccan_Jews look like that in terms of autosomal DNA. This is why we must be careful with these tools:

the fits are bad for everyone so it could be correct. it's maybe just that all other populations are so far away from them now so that jewish populations are closest.
the thing is maybe that the "levantine farmer" ancestry is not differentiated strong enough from anatolian farmer/CHG to pull jewish people, who also have anatolian_N/CHG further away than the other populations who also have different admixtures.

 

[blevins13]

Dodecad Globe 13 for Mik15 vs available modern populations:

Distance to:	Mik15
14.79754372	Morocco_Jews
16.63085386	Sephardic_Jews
17.52413479	Sicilian_D
17.78300593	S_Italian_Sicilian_D
19.19753370	Ashkenazi_D
19.37811394	C_Italian_D
19.41410055	Ashkenazy_Jews
22.80840415	Tuscan
23.31169878	TSI30
23.33840826	Moroccan_D
24.24646985	Greek_D
25.28409184	Cypriots
25.50218226	Sardinian
25.87959235	Algerian_D
26.61235240	Canarias_1KG
26.75319233	Mozabite
26.88983637	North_Italian
27.19590594	Turkish_Cypriot_D
27.83072583	O_Italian_D
29.10995191	Andalucia_1KG
29.12918296	Moroccans
29.31107811	Murcia_1KG
30.27208780	N_Italian_D
31.58302234	Baleares_1KG
31.68168714	Portuguese_D

Here is Mik15 vs the other ancient Greek samples:

He is very close to Minoans, which we know are mostly Anatolian_N with about 15% CHG/IN. There is no way Algerian_Jews are composed of that admixture. We can't really compare these samples so much to modern populations, because the inferences made by doing so are bound to be false:

Distance to:	Mik15
4.73667605	I2937
5.72554801	I9129
5.90445594	I9010
7.59130424	Pta08
7.62219785	I9128
9.00139989	I9005
9.35317593	I0074
9.42462731	Kou01
9.80234156	I0071
11.08796194	I0073
11.33379460	I0070
11.56451901	I9130
12.19090235	I9127
12.45656453	I9131
13.06872985	I9041
13.70391915	I9006
14.52259274	I9033
15.10968564	Kou03
18.59169438	I9123
18.99046866	I2495
22.03010667	I2499
22.92758600	I2683
25.84222707	Log02
34.33169236	Log04

Autosomal-wise Mik15 looks like a Minoan, with an additional small amount of Steppe. Again, there is no way Algerian_Jews or Moroccan_Jews look like that in terms of autosomal DNA. This is why we must be careful with these tools:

All Minoan samples yield similar results.

Minoan Moni Odigitira 2000 BC. This is known from 2017, looks reasonable to me considering location and relations over time. I don’t see anything misleading in these cases. Certainly there are other cases that might be misleading, so ingredients should be checked.

Closest Modern.

1. Algerian_Jewish (10.90)
2. Italian_Jewish (12.46)
3. Tunisian_Jewish (13.79)
4. Libyan_Jewish (15.23)
5. South_Italian (16.78)
6. Maltese (17.77)
7. West_Sicilian (18.49)
8. East_Sicilian (19.64)



Sent from my iPhone using Eupedia Forum

 

[Archetype0ne]

Do we have ancient Jewish autosomal DNA?
Were they mostly Anatolian_N ? How would they compare to these Minoan samples? What about Phoenician autosomal (they were moving around, and could have been a vector of transmission)?

 

[Stuvanè]

On the other hand, I'm rather inclined to believe that the peoples from which Log02 and Log04 originate are not the true carriers of the ancient Greek language (or that they are secondary). More likely they are the carriers of that portion of steppic DNA that later diluted into the Minoan substrate, giving rise to the Mycenaeans proper a few centuries later.
I'd be more willing to bet that the bulk of the Mycenaean language (and culture) is probably an inheritance that came mainly from Anatolia/Caucasus, perhaps on a par with the language of the Minoans (already several scholars had gone so far as to speculate that there were extremely ancient Caucasian or Anatolian Indo-European languages behind the mysterious Linear A itself). I'm thinking of various statistical and glottochronological models according to which the Greek language, although much later, still remains quite close to that of the Hittites, which at the time is considered the most archaic of the family, and in any case quite distinct from the superclade of other linguistic families carried into continental Europe by the Yamnaya and their Corded-Ware or related heirs.


This may imply, if we want to save the day, that the Indo-Europeanisation of the archaic Greek world is perhaps the result of two combined directions (Eastern Model + Northern Model mentioned by Lazaridis a few years ago), overlapping and alternating over centuries or even millennia, which like a pincer imposed itself and infiltrated the Neolithic and Minoan substrate.


Warning: I'm NOT referring to Renfrew's old hypothesis about the Indo-Europeanisation of the continent through Neolithic expansion, but to an archaic Indo-European proto-component different from the steppic one and initially distinct from the Neolithic farmers - composed mostly of CHG - and located perhaps between the inner Caucasus and Iran that continued to operate and ravage the region in a substantially autonomous way from the steppe peoples (a part of this component in earlier periods joined with the EHG to form the Yamnaya proper).


It is a hypothesis that could hold together the linguistic and genetic data (an Indo-European language showing some important divergences from the other families + ancient Indo-European peoples of the Anatolian and Aegean region who at the time of their peak were characterised by a very modest genetic signal from the steppe)

 

[Jovialis]

the fits are bad for everyone so it could be correct. it's maybe just that all other populations are so far away from them now so that jewish populations are closest.
the thing is maybe that the "levantine farmer" ancestry is not differentiated strong enough from anatolian farmer/CHG to pull jewish people, who also have anatolian_N/CHG further away than the other populations who also have different admixtures.

I don't have Algerian_Jews, but Moroccan_jews are quite different from minoans in autosomal DNA

Minoans are mostly Anatolian_N with a bit of CHG, as demonstrated by the study of this thread.

Moroccon_Jews are as such:

They are different too, Natufian is very different from CHG as demonstrated by PCA analysis:

Also there are other exotic admixture in them which make Moroccan_Jews different as well.

 

[Jovialis]

What makes Natufians different primarily is 27% Talforat-like (Ancient North African) admixture:

 

[Archetype0ne]

On the other hand, I'm rather inclined to believe that the peoples from which Log02 and Log04 originate are not the true carriers of the ancient Greek language (or that they are secondary). More likely they are the carriers of that portion of steppic DNA that later diluted into the Minoan substrate, giving rise to the Mycenaeans proper a few centuries later.
I'd be more willing to bet that the bulk of the Mycenaean language (and culture) is probably an inheritance that came mainly from Anatolia/Caucasus, perhaps on a par with the language of the Minoans (already several scholars had gone so far as to speculate that there were extremely ancient Caucasian or Anatolian Indo-European languages behind the mysterious Linear A itself). I'm thinking of various statistical and glottochronological models according to which the Greek language, although much later, still remains quite close to that of the Hittites, which at the time is considered the most archaic of the family, and in any case quite distinct from the superclade of other linguistic families carried into continental Europe by the Yamnaya and their Corded-Ware or related heirs.


This may imply, if we want to save the day, that the Indo-Europeanisation of the archaic Greek world is perhaps the result of two combined directions (Eastern Model + Northern Model mentioned by Lazaridis a few years ago), overlapping and alternating over centuries or even millennia, which like a pincer imposed itself and infiltrated the Neolithic and Minoan substrate.


Warning: I'm NOT referring to Renfrew's old hypothesis about the Indo-Europeanisation of the continent through Neolithic expansion, but to an archaic Indo-European proto-component different from the steppic one and initially distinct from the Neolithic farmers - composed mostly of CHG - and located perhaps between the inner Caucasus and Iran that continued to operate and ravage the region in a substantially autonomous way from the steppe peoples (a part of this component in earlier periods joined with the EHG to form the Yamnaya proper).


It is a hypothesis that could hold together the linguistic and genetic data (an Indo-European language showing some important divergences from the other families + ancient Indo-European peoples of the Anatolian and Aegean region who at the time of their peak were characterised by a very modest genetic signal from the steppe)

Check this out Stuvane. https://www.eupedia.com/forum/threa...vid-Goldstein-Indo-European-Linguistics-V8-I1

 

[Archetype0ne]

On the other hand, I'm rather inclined to believe that the peoples from which Log02 and Log04 originate are not the true carriers of the ancient Greek language (or that they are secondary). More likely they are the carriers of that portion of steppic DNA that later diluted into the Minoan substrate, giving rise to the Mycenaeans proper a few centuries later.
I'd be more willing to bet that the bulk of the Mycenaean language (and culture) is probably an inheritance that came mainly from Anatolia/Caucasus, perhaps on a par with the language of the Minoans (already several scholars had gone so far as to speculate that there were extremely ancient Caucasian or Anatolian Indo-European languages behind the mysterious Linear A itself). I'm thinking of various statistical and glottochronological models according to which the Greek language, although much later, still remains quite close to that of the Hittites, which at the time is considered the most archaic of the family, and in any case quite distinct from the superclade of other linguistic families carried into continental Europe by the Yamnaya and their Corded-Ware or related heirs.


This may imply, if we want to save the day, that the Indo-Europeanisation of the archaic Greek world is perhaps the result of two combined directions (Eastern Model + Northern Model mentioned by Lazaridis a few years ago), overlapping and alternating over centuries or even millennia, which like a pincer imposed itself and infiltrated the Neolithic and Minoan substrate.


Warning: I'm NOT referring to Renfrew's old hypothesis about the Indo-Europeanisation of the continent through Neolithic expansion, but to an archaic Indo-European proto-component different from the steppic one and initially distinct from the Neolithic farmers - composed mostly of CHG - and located perhaps between the inner Caucasus and Iran that continued to operate and ravage the region in a substantially autonomous way from the steppe peoples (a part of this component in earlier periods joined with the EHG to form the Yamnaya proper).


It is a hypothesis that could hold together the linguistic and genetic data (an Indo-European language showing some important divergences from the other families + ancient Indo-European peoples of the Anatolian and Aegean region who at the time of their peak were characterised by a very modest genetic signal from the steppe)

Very well though out post by the way.


Figure 25

MLE tree with bootstrap scores (NJ start tree)
Citation: Indo-European Linguistics 8, 1 (2020) ; 10.1163/22125892-20201000

Your theory adds ideas in my mind about possible early Indo European expansions. A not yet published possible l283 sample in Moldova could be as old as 6000 years, since it is speculated it is from the Eneolithic. Which for my own clade changes the whole picture. Albeit we have to wait for them to publish such a paper after 2 years of the leak.

Considering the old basal L283 are found around North Caucasus, next to the steppes, as well as considering L283* is found in a Modern Armenian, and J was around Zargos early on, you theory in my mind has some circumstantial, albeit weak support.

 

[Jovialis]

Do we have ancient Jewish autosomal DNA?
Were they mostly Anatolian_N ? How would they compare to these Minoan samples? What about Phoenician autosomal (they were moving around, and could have been a vector of transmission)?

The study of which this thread is based on has dismissed this possiblity:

[FONT=&quot]Other proposed migrations, such as settlement by Egyptian or Phoenician colonists[/FONT][FONT=&quot]22[/FONT][FONT=&quot] are not discernible in our data, as there is no measurable Levantine or African influence in the Minoans and Myceneans, thus rejecting the hypothesis that the cultures of the Aegean were seeded by migrants from the old civilizations of these regions.

[/FONT]https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5565772/

 

[Archetype0ne]

The study of which this thread is based on has dismissed this possiblity:

Which possibility?

Oh I just saw the quote.

Yeah I wasn't talking about that. Was asking what Ancient Jews and Phoenician auDNA composition is.