Anatolia
has witnessed many migrations and has been home to a variety of civilizations.
The increasing availability of ancient DNA samples from different periods and
cultures makes it now possible to examine the genetic changes that have taken
place in Anatolia since the Neolithic. The ancient DNA revolution has already dispelled
many myths, but myths and legends die hard. Some of the more persistent ones concern
the impact of the last migration wave to hit Anatolia, marked by the arrival of
the Oghuz Turks. While the Oghuz’ cultural impact is undeniable, the
significance of their genetic legacy is still contested by some. Although there
are still some missing links in the DNA record, the available evidence allows
us to put the “no true Turkmen fallacy” to the test.
Anatolia_Neolithic: Neolithic sample from central Anatolia (Tep002, Tepecik-Çiftlik, 6680-6590 BC).
Anatolia_Chalcolithic: Chalcolithic sample from western Anatolia (I1584, Barcın, 3943-3708 BC).
Anatolia_Bronze_Age: Bronze Age sample from western Anatolia (I2683, Harmanören-Göndürle, 2500-1800 BC).
Greek_Cappadocia: Average of Cappadocian Greek samples from our project databese.
Turk_Cappadocia: Average of Cappadocian Turkish samples from our project databese.
Turk_Anatolia: Average of all Anatolian Turkish samples from our project database.
Due to its inclusion of hunter-gatherer and farmer related components, I've thought it would be appropriate to start with the puntDNAL K12 calculator.
| Tep002 | I1584 | I2683 | ||||
|---|---|---|---|---|---|---|
PUNTDNAL K12
|
Anatolia_Neolithic
|
Anatolia_Chalcolithic
|
Anatolia_Bronze_Age
|
Greek_Cappadocia
|
Turk_Cappadocia
|
Turk_Anatolia
|
Sub-Saharan
|
-
|
0.34
|
2.62
|
-
|
0.34
|
0.28
|
Amerindian
|
-
|
-
|
0.55
|
-
|
0.85
|
0.73
|
South_Asian
|
-
|
-
|
1.03
|
0.91
|
3.45
|
3.43
|
Near_East
|
-
|
6.71
|
17.18
|
15.04
|
11.8
|
11.48
|
Siberian
|
-
|
0.46
|
-
|
0.47
|
4.75
|
5.56
|
European_HG
|
-
|
7.36
|
5.73
|
8.24
|
10.45
|
10.56
|
Caucasus_HG
|
-
|
40.93
|
28.68
|
36.03
|
34
|
33.85
|
South_African_HG
|
-
|
-
|
0.2
|
-
|
0.36
|
0.28
|
Anatolia_NF
|
100
|
44.21
|
43.86
|
37.94
|
29.08
|
28.46
|
East_Asian
|
-
|
-
|
-
|
-
|
3.36
|
3.61
|
Oceanian
|
-
|
-
|
0.15
|
0.79
|
0.59
|
0.67
|
Beringian
|
-
|
-
|
-
|
0.59
|
0.96
|
1.1
|
Neolithic to Chalcolitic transition: We see a dramatic decrease in the Anatolian_Neolithic component in the Chalcolithic sample, and an increase in the Caucasus_HG component. This change seems to be caused by a large-scale migration of people heavy in Caucasus_HG/Iran_N ancestry.
Chalcolithic to Bronze Age transition: The low quality of this sample causes the admixture results to be quite noisy, but an increase in the Near_East component can be observed. Lazaridis 2016 determined that the Bronze Age sample can be best modelled as a three-way mix of 62% Anatolia_N + 32% CHG + 6% Levant_N.
Let's continue with the Eurogenes K13 calculator.
| Tep002 | I1584 | I2683 | ||||
|---|---|---|---|---|---|---|
| EUROGENES K13 |
Anatolia_Neolithic
|
Anatolia_Chalcolithic
|
Anatolia_Bronze_Age
|
Greek_Cappadocia
|
Turk_Cappadocia
|
Turk_Anatolia
|
North_Atlantic
|
12.94
|
13.04
|
5.14
|
7.15
|
7.23
|
7.8
|
Baltic
|
-
|
-
|
-
|
3
|
6.63
|
6.7
|
West_Med
|
33.33
|
28.21
|
24.84
|
19.3
|
11.19
|
11.49
|
West_Asian
|
1.17
|
24.12
|
23.58
|
29.63
|
28.66
|
28.23
|
East_Med
|
39.79
|
34.63
|
40.15
|
34.89
|
28.22
|
26.85
|
Red_Sea
|
9.09
|
-
|
5.89
|
5.18
|
4.26
|
4.32
|
South_Asian
|
-
|
-
|
-
|
0.04
|
3.43
|
3.32
|
East_Asian
|
-
|
-
|
0.4
|
0.07
|
3.17
|
3.21
|
Siberian
|
0.45
|
-
|
-
|
0.59
|
5.62
|
6.46
|
Amerindian
|
-
|
-
|
-
|
-
|
0.7
|
0.63
|
Oceanian
|
-
|
-
|
-
|
0.18
|
0.43
|
0.45
|
Northeast_African
|
2.74
|
-
|
-
|
-
|
0.21
|
0.3
|
Sub_Saharan
|
0.52
|
-
|
-
|
-
|
0.21
|
0.23
|
If we assume that our modern-day Cappadocian Greek average is representative of the genetic profile of Cappadocian Greeks during the times of the Byzantine Empire, then we can conclude that the Oghuz migrations, which started in the 11th century and continued for hundreds of years, had a significant impact on the genetic structure of Anatolia.
The differences we observe between our samples of Cappadocian Greeks and Cappadocian Turks allows us to gain some insights into the genetic profile of the incoming Oghuz Turks:
- From the observation that Anatolia_Neolithic related components (such as West_Med and East_Med) show significant declines, while the West_Asian component doesn't, we can deduce that this component was one of the major components that the Seljuk era Oghuz brought with them.
- The increase that we can see in the Baltic component is not surprising, given that modern Central Asians have greater amounts of Baltic admixture than West Asians.
- The fact that the proportions of the West Asian component barely changed, while at the same time the South_Asian component (a proxy for Iran_N related ancestry) increased, indicates that we should focus on the region south of the Aral Sea.
Let's add a couple of samples from the south of the Aral Sea.
Turkmen_TM: Average of Turkmen samples from Turkmenistan (Ashgabat).
Turkmen_UZ: Average of Turkmen samples from Uzbekistan.
Uzbek_UZ: Average of Uzbek samples from Uzbekistan.
| I2683 | |||||||
|---|---|---|---|---|---|---|---|
EUROGENES K13
|
Anatolia_BA
|
Greek_Cappadocia
|
Turk_Cappadocia
|
Turk_Anatolia
|
Turkmen_TM
|
Turkmen_UZ
|
Uzbek_UZ
|
North_Atlantic
|
5.14
|
7.15
|
7.23
|
7.8
|
6.39
|
7.25
|
5.2
|
Baltic
|
-
|
3
|
6.63
|
6.7
|
7.41
|
9.48
|
13.28
|
West_Med
|
24.84
|
19.3
|
11.19
|
11.49
|
2.84
|
3.13
|
1.76
|
West_Asian
|
23.58
|
29.63
|
28.66
|
28.23
|
34.2
|
25.95
|
24.43
|
East_Med
|
40.15
|
34.89
|
28.22
|
26.85
|
17.38
|
11.85
|
5.67
|
Red_Sea
|
5.89
|
5.18
|
4.26
|
4.32
|
2.38
|
3.01
|
1.41
|
South_Asian
|
-
|
0.04
|
3.43
|
3.32
|
10.79
|
9.77
|
9.15
|
East_Asian
|
0,4
|
0.07
|
3.17
|
3.21
|
6.18
|
10.3
|
14.08
|
Siberian
|
-
|
0.59
|
5.62
|
6.46
|
10.1
|
17.5
|
22.66
|
Amerindian
|
-
|
-
|
0.7
|
0.63
|
1.1
|
0.98
|
1.32
|
Oceanian
|
-
|
0.18
|
0.43
|
0.45
|
0.47
|
0.37
|
0.25
|
Northeast_African
|
-
|
-
|
0.21
|
0.3
|
0.21
|
0.29
|
0.47
|
Sub_Saharan
|
-
|
-
|
0.21
|
0.23
|
0.55
|
0.12
|
0.32
|
We can take a look at the closest populations for some of our sample averages with the help of the 4-Ancestors Oracle. The numbers to the right of the population names (@ x) are least-squares distances, the smaller the numbers the closer the populations are to each other. A distance greater than 10 cannot be considered close.
Anatolia_Chalcolithic
Least-squares method.
Using 1 population approximation:
1 Anatolia_Chalcolithic @ 0
2 Anatolia_Bronze_Age @ 11,805973
3 Greek_Cappadocia @ 13,443617
4 South_Italian @ 14,011506
5 Cyprian @ 15,958764
6 Sephardic_Jewish @ 17,037429
7 Central_Greek @ 17,24864
8 Italian_Jewish @ 17,53137
9 East_Sicilian @ 17,655078
10 Italian_Abruzzo @ 19,476547
211 iterations.
Anatolia_Bronze_Age
Least-squares method.Anatolia_Chalcolithic
Least-squares method.
Using 1 population approximation:
1 Anatolia_Chalcolithic @ 0
2 Anatolia_Bronze_Age @ 11,805973
3 Greek_Cappadocia @ 13,443617
4 South_Italian @ 14,011506
5 Cyprian @ 15,958764
6 Sephardic_Jewish @ 17,037429
7 Central_Greek @ 17,24864
8 Italian_Jewish @ 17,53137
9 East_Sicilian @ 17,655078
10 Italian_Abruzzo @ 19,476547
211 iterations.
Anatolia_Bronze_Age
Using 1 population approximation:
1 Anatolia_Bronze_Age @ 0,453767
2 Cyprian @ 7,171295
3 Greek_Cappadocia @ 10,516959
4 Anatolia_Chalcolithic @ 11,841659
5 Lebanese_Muslim @ 12,632653
6 Lebanese_Christian @ 14,350529
7 Samaritan @ 14,435627
8 Syrian @ 15,324154
9 Sephardic_Jewish @ 16,596795
10 Armenian_West @ 17,191345
211 iterations.
Greek_Cappadocia
Least-squares method.
Using 1 population approximation:
1 Greek_Cappadocia @ 0,245893
2 Armenian_West @ 10,245416
3 Anatolia_Bronze_Age @ 10,41903
4 Greek_Pontus @ 10,828287
5 Cyprian @ 11,707179
6 Assyrian @ 12,161285
7 Lebanese_Muslim @ 12,184882
8 Turk_Cappadocia @ 13,231142
9 Anatolia_Chalcolithic @ 13,441605
10 Turk_Anatolia @ 14,188997
211 iterations.
We can make the differences clearer with the help of a PCA of Anatolia and Central Asia. The Turkish average is located between Greek_Cappadocia and Central Asian averages. The Central Asian genetic legacy of Turks pulls Turk_Anatolia away from Greek_Cappadocia, Anatolia_Bronze_Age and Anatolia_Chalcolithic.
In light of all this data, we can conclude that during the long time span from the Neolithic to the Oghuz migrations the genetic structure of Anatolia underwent successive transformations, and that the last big change was caused by the Oghuz migrations.
SOURCES:



Very interesting, really good work and good sources. Love the attention to acedemia!
ReplyDeleteSo the noticeable difference between Greek and Turk Capadocian is essentially:
* +3% Baltic
* -8% West_Med
* -5% East_Med
* 3% South_Asian
* 3% East Asian
* 5% Siberian
Which would be a total of approximately 27%, could one from these results and differences conclude that the amount of Turkic migrators to the Capadocia region was 27% of the Capadocian population? Since they made a 27% difference in the Genepool.
Another thing I'm concerned about is if any other historical migrations and/or raids are taken into consideration? Are we sure the -8% West_Mediterian isn't simply from more Iranians raiding/settling some parts of Capadocia?.
For example we know that the Capadocians even elected persian lords(Ariobarzanes). They got dethroned and probably raided/settled by Armenians in 63 BC. And I'm not even an historian, I'm sure someone else could dig alot of more pre-turkic conflicts from that region that might have changed the Greek Capadocian genepool before Turks entered that region.
I REALLY enjoy your posts, please keep up the good work.