Genetic history of Levant. Part A.

 Genetic history of Levant. Part A.

Recently there was a surge of interest in Levantine genetics, so here is a summary of Levant genetic history in two parts

Natufian culture (13000-9000BC) a hunter gatherer society which has a remarkable feature of being sedentary or semi sedentary. Archaic E1b and CT haplogroups.

Levant Neolithic (9000-4500BC) is divided in two parts. Pre Pottery Neolithic (PPN) and Pottery Neolithic. Spread of farming to south Levant from more northern regions. The core region of farming invention was in Urfa region ( Portasar / Gobekli tepe ) near modern Syrian Turkey border. You can see on the chart an important shift to north in autosomes during this period in south Levant ( Israel , Jordan PPNB ). A best proof of this scenario is the haplotype E1b-Z1919 which has two important branches. One E-L618 moved to Europe via Anatolia and today is popular in Balkans. The other parallel branch E-V22 moved to Egypt introducing farming there. It is popular today in north Egypt. Another important lineage from this period is the E-M123 mostly found today in West Asia even though it moved to Central Asia also. Sporadic apparition of H2 and T1a.

Copper Age - Chalcolithic ( 4500-3300BC) there are no much samples from this period. The few cases from a period close to Ghassulian culture show another shift in autosomes toward north Iraq. Consistent with the archaeological records showing emergence of new burial tradition. Currently it's hard to pinpoint the exact cultural phenomenon that triggered this shift in Chalcolithic south Levant. The similar shift toward north Mesopotamia started in north Levant (Tell Kurdu) earlier. The plausible candidate is the Halaf culture. The T1a1a-L208 in this period is also favoring the theory of involvement of new groups not directly descending from Neolithic period.

Early Bronze Age (3300-2200BC) the end of Chalcolithic and the start of Bronze Age in Levant is a period of dramatic changes in material culture. Old settlements are abandoned and new mobile pastoralist groups spread all over Levant. Those are the early Semitic tribes. The change in autosomes is subtle, more eastern but the change in Y DNA in Levant EBA is drastic. The E1b shrinks and the J1-Z1853 is the most popular haplotype starting from this period. J2b1 also shows an important increase in this period. Where was exactly located the Proto Semitic homeland is debated. It is usually placed in Arabian peninsula. The slight increase of Natufian ancestry do favour a southern location. But the surge of ancestry related to historic Armenia force to imagine a region close to Mesopotamia (see the map) A region where J1 must have been present since older periods. J1 is ultimately from more northern Taurus-Zagros-Caucasus mountains.

Currently the ancient DNA is insufficient to locate precisely the Proto Semitic homeland and the path of J1 migration.

Wherever the Proto-Semitic homeland was located they expanded to north reaching Taurus mountains and to east reaching Zagros mountains. In the south they crossed Red sea and landed in Ethiopia around 1000BC. The first branch in Semitic family is the East Semitic represented by Eblaite and Akkadian. Both extinct today. They settle in north Levant and Mesopotamia where their culture merged with the Sumerians and apparently put end to Uruk culture at 3100BC. Akkadians created one of oldest empires, uniting Mesopotamian city states around 2330BC which collapsed around 2150BC.

After the 2200BC climatic event and the demise of Akkadian empire new period starts in Levant which will be discussed in Part B

Alashkert Armenians have quite remarkable genetics.

Alashkert Armenians have quite remarkable genetics. They are the closest in fst runs to three geographically distant from each other groups. Artsakh, Sasun and Hamshen Armenians.

Is this due to their central "average" position in the Armenian regional diversity or has some deeper reasons is hard to say at this stage.

Negative fst value means zero. Which means lack of any meaningful distance.

Genetic history of Greece

 Genetic history of Greece

Yediay et al. 2024 published large number of ancient DNA from Bronze Age Greece which permitted to solve the question of Greeks origin. If we skip the unsampled Paleolithic period Greece has witnessed three important migrations in prehistoric times.

+ Neolithic farmers from Anatolia, ultimately from Urfa region moved to Greece around 7000 BC. Farming was introduced earlier in Greece than in any part of Europe. From there farmers moved to north.

+ Chalcolithic migrants from historic Armenia around 4000-3000 BC. This second migration was associated with Chaff faced ware in Anatolia. While in Greece it left the strongest impact in Crete and Aegean Cycladic islands. It can be connected to the formation of Minoan civilization. (Linear A) Even though Heggarty 2023 wanted to link it with Greeks.

This migration introduced new haplotypes and in most likelihood is partly responsible for the so called Mediterranean substrate present both in Greek and Armenian. An examlle is the iwl, el (իւղ, եղ) meaning oil in Armenian. Compare to Greek elaiwon (ἔλαιον) olive oil.

+ Migration from Pontic Caspian Steppe around 2200 BC. Similar to Armenia the change from Early Bronze Age to Middle Bronze Age is associated with migrations from Pontic Caspian steppe. Those migrants introduced haplotypes like R1b-Z2103 and R1b-PF7562. With few cases of I2 and J2b2a. Mycenaean Greeks also had large number of J2-Z6055 a lineage that was present in Greece since the Neolithic period. High level of that lineage are probably due to founder effect.

After a short MBA period there is a dilution of steppe ancestry in Late Bronze Age Greece (1700-1100BC). A period associated with Mycenaean culture. (See the chart) Linear B has been deciphered and linked to early or parallel extinct forms of Greek. Mycenaean Greeks had ancestry both from local Neolithic farmers and preceding migrants from more eastern regions. In the chart it's marked as Anatolia C. Which is related to the CFW from historic Armenia.

Yediay et al. 2024 linked the steppe migration with Indo Europeans from Yamnaya culture. Thus solving in convincing manner the origins of Greek people.

Around 1200 BC the Mycenaean civilization collapsed and many people were forced to emigrate. During this period number of groups moved to eastern Mediterranean where they were known as Sea people. One of those groups settled in Cyprus introducing Greek language there.

Even prior that Greeks were already settling in Western Anatolian coasts. In that region they are known probably as Ahhiyawa in Hittite texts. Multiple wars started for the control of western Anatolian coasts. The memory of those wars is preserved in the Greek epic poems known as Illiad and Odyssey.

Modern Greek is usually assumed to be derived from Attic-Ionic dialect

Attempt to redo the models from recent papers.

 Attempt to redo the models from recent papers.

Armenia LBA as a simple mixture of Kura-Araxes and Catacomb is a failure. P value too low. ( pic. 1)
Addition of an Anatolian population like Ikiztepe C improves the p value. This was done in Skourtanioti 2024. The model is feasible. p=0.2
The use of CamlibelTarlasi_LateC instead of Ikiztepe improves the p value even more. Done in Yediay 2024. p more than 0.3
An addition of Leila tepe culture (from what is now Azerbaijan) instead of Anatolian C improves the p value much more. p more than 0.7. But the standard errors increase a lot because the Kura-Araxes and Leila tepe are too close genome wide to each other. To reduce the standard errors I need to change the settings (right pops). This will require a time from me. But I think its quite obvious what is going here.
And finally I add the Ukraine N hunter gatherers to the source and the p value reaches practically 0.9, close to certitude. The standard errors also improves slightly, but still requires improvement.

Conclusions. The populations that played a role in the formation of Trialeti Vanadzor culture came from what is now the Krasnodar region, acquired there a minor Ukraine_N HG ancestry. Moved via Dagestan mixed with older South Caucasian Chalcholithic groups in Kura-Araxes plains then splits into two branches. One moved to upstream to Kur-Debed region and created the TVC, while the other moved to Urmia and Van basins and created Van-Urmia culture even though their genetic impact there was lower than in TVC. The Y DNA associated with those people should be searched in Krasnodar region. L584, I2a2b, PH331. Finding Y4364 will be harder. Its even possible that the true homeland of Proto-Yamnaya was also there

Before I review the recent Yedia et al. 2024 preprint few comments are needed about this chart.

 Before I review the recent Yedia et al. 2024 preprint few comments are needed about this chart.

Armenia Middle and Late Bronze Age samples do have ancestry from three sources. The first one is from Armenia EBA the Kura-Araxes culture which is expected and do not require an explanation.

The second stream of ancestry is marked as Anatolia_C. This could be surprising but has an easy explanation. Anatolia_C is largely derived from historic Armenia Neolithic and Chalcolithic groups known as Chaff faced ware. Those groups lived before the Kura-Araxes and where present in modern Azerbaijan where they were known as Leila tepe culture. When Caucasian HG shifted Kura-Araxes expanded from it's homeland in South Caucasus it didn't replaced all those preceding people and in many places they continued to live. When 4500 years ago steppe ancestry came from north they first crossed what is now Azerbaijan and in most likelihood encountered the remnants of Leila tepe culture. They mixed with them acquired the "Anatolia_C" like ancestry and then moved to upper sources of Kur-Debed rivers and mixed with the Armenia EBA related people. Another source of Anatolia_C like ancestry could be from southern regions of Arax plain.

This by the way was noted by Skourtanioti 2024 also but they gave a wrong explanation thinking that there was a two separate migrations. One from Anatolia and another from Steppe during the same period. Genetic bloggers have noticed this much earlier and Davidski has even a thread dedicated to it.

And finally the presence of so called CWC ancestry. It is obviously not real because CWC is highly associated with R1a and there was no any R1a in Middle and Late Bronze Age South Caucasus ( from Armenia and Georgia). The reason why CWC appears there is because the group of steppe people that moved to South Caucasus most probably had some WHG affinity. This extra WHG/UNHG affinity was present in east of Azov region even before Yamnaya formed and most probably is the plausible explanation of the apparition of I2a2b in ancient Armenia. Notice that CWC is similar to Yamnaya but has extra 10% UNHG related ancestry and some 20% of Euro-Anatolian farmer ancestry. Even a small 1% introgression of WHG into a Yamnaya population will create an effect of presence of 10% CWC ancestry and given the excess of Anatolian ancestry in MLBA it's not surprising that the calculator assumed that he deals with CWC rather than a Yamnaya. So this is without doubt an artefact of calculation which nevertheless has its objective causes.

The same can be said about Urartu samples where the Anatolia_C is higher, thus the ancient Neolithic ancestry is higher while steppe ancestry is lower. Except one outlier who was a migrant from Etiuni

One of Greek samples from Hovhannisyan 2024 is located in the modern Armenian cloud.

 One of Greek samples from Hovhannisyan 2024 is located in the modern Armenian cloud. I couldn't find any information about him except that Greek DNA used in the paper was from Lazaridis 2014 Nature.

Based on his location on the PCA he is not a Cappadocian Greek. Also he do not give impression to be related to Trabzon Greeks. Most probably he is from Urums from South West of Georgia. Urums settled in Tsalka region of Georgia after Russian and Turkish wars of 19th century.

We have an Urum person in our group and his DNA could also be similar to that person. Off course this is a just an assumption that needs further verification.

Urums should not be confused with Urumu tribe of Iron Age. The word Urum is derived from Rome, Roman empire. While the name of Urumu has unknown origin, but could be well related to Aramu.

In the second PCA You can see the main modern Greek cloud which is away from modern Armenians. It is somewhat pulled toward north compared to Mycenaean period. Another Greek samples is almost certainly from Anatolia. Located in the middle of Armenians and Greeks.