The genomic history of the Aegean palatial civilizations.

Status
Not open for further replies.
Angela said:
88a4060812e389aab3f69d9bca5c2f9a.jpg


This is my husband's complexion, and how deeply he can tan, while I huddle under an umbrella and can still wind up with skin poisoning. The only important thing about skin ***mentation is whether you're adapted to your climate. I seem to be adapted for very northern latitudes or high atop some mountains; I just don't like them. :)
Click to expand...

The actor in the last picture is likely how the ancient Greeks looked like in terms of complexion. It’s a bit odd, to think that a 5000 years old predynastic Egyptian that was naturally mummified and exposed to the heat and sun, is still visibly, pretty light skinned. Which means BA Greeks were big time darker than the predynastic Egyptians.​
38262_17913_228099568.jpg
 
iluvatar said:
Log4 doesn’t cluster anywhere near Scotts, she clearly clusters near south Balkan populations (even if, and to some extent definitely because of coincidence)

F3 stats dont measure overall similarity. Notice how Anatolia bronze age is mostly similar to Sardinians going by their f3 chart, but on any pca bronze age Anatolians will cluster far away from them and closer to populations like Cypriots. Same with the Myceneans, it's just that both populations have Anatolia Neolithic as their predominant component.
Click to expand...

Every tool has its good points and its bad points. That's why they should all be looked at. Population Genetics is not for linear thinkers.

PCAs show only two dimensions and a reduced percentage of the total genes. Novembre et al, for example, captured 45% of them. Some capture far fewer.

https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1002886

I shouldn't need to point out that landing near one another on a PCA doesn't necessarily mean that the two populations share the same population history.

As to F3 etc., perhaps a review of some of the statistical tools is in order.

https://compvar-workshop.readthedoc...duced by Nick Patterson in his Patterson 2012
 
real expert said:
The actor in the last picture is likely how the ancient Greeks looked like in terms of complexion. It’s a bit odd, to think that a 5000 years old predynastic Egyptian that was naturally mummified and exposed to the heat and sun, is still visibly, pretty light skinned. Which means BA Greeks were big time darker than the predynastic Egyptians.​
38262_17913_228099568.jpg
Click to expand...

R.E. they're all the same man. My point was that people who look "white" in winter, can get much darker in summer.

I don't know whether the "Griffin Man" was naturally that dark or it was a tan.
 
Angela said:
R.E. they're all the same man. My point was that people who look "white" in winter, can get much darker in summer.

I don't know whether the "Griffin Man" was naturally that dark or it was a tan.
Click to expand...

I also don't think your last conclusion was warranted.

Egyptians knew very well how they looked in comparison to Sudanese people and Semites, much less Greeks.

A-Nubian%2C-a-Syrian%2C-and-an-Egyptian.jpg


If t-rolling is going to start I'm going to stop responding. PLEASE read the paper and all the Supplement. All I'm seeing is a lot of misinformation from other sites etc.

One simple thing: Only Log 4 is about 50% steppe. He probably just arrived, for goodness' sakes. Log 2 is very different.
 
Angela said:
R.E. they're all the same man. My point was that people who look "white" in winter, can get much darker in summer.

I don't know whether the "Griffin Man" was naturally that dark or it was a tan.
Click to expand...

I can vouch for that. Due to my 10% Sicilian DNA, I tan quite well; as a kid, a March picture of me looked like a fine English boy, while in a September shot, I could have passed for Hispanic. Even today I have a semipermanent 'farmer tan'.
 
Angela said:
I also don't think your last conclusion was warranted.

Egyptians knew very well how they looked in comparison to Sudanese people and Semites, much less Greeks.

A-Nubian%2C-a-Syrian%2C-and-an-Egyptian.jpg


If t-rolling is going to start I'm going to stop responding. PLEASE read the paper and all the Supplement. All I'm seeing is a lot of misinformation from other sites etc.

One simple thing: Only Log 4 is about 50% steppe. He probably just arrived, for goodness' sakes. Log 2 is very different.
Click to expand...


Angela, you got wrong what my point was, probably due to the fact, that you reply to several people. Firstly, I know that you showed me the same guy, Raoul Bova who was shown on this forum several times. I took his tanned complexion to make clear that "dark" in Greek context means such a skin tone and not SSA dark. Plus, I wanted to point out, if the prediction for the two samples is accurate and not flawed, that would mean that these tested Aegeans were darker than this predynastic Egyptian who was clearly not dark brown till black but moderately light skinned. So it‘s beyond me why you think I‘m tr*lling and accuse me of not being interested on your take on this study. If that wasn't the case I wouldn't ask and discuss with you. Again, I was referring to the prediction which we all find surprising because nobody expected very dark to black ancient Greeks as this study proposes. However, we know from history and Greek sources that Ancient Greeks considered Egyptians darker than them. Besides, I talked about the steppe admixed sample since her phenotype was predicted.
 
I shouldn't need to point out that landing near one another on a PCA doesn't necessarily mean that the two populations share the same population history.
Click to expand...

Obviously, but I never claimed that modern south balkanites are unadmixed descendants of these samples. Clearly later movements were involved as well as absorption of natives, evident in the fact that these samples score very little CHG/IRAN Neo while both Minoans and Cycladic/Helladic EBA do.

A thing to note is that their pca is way off. Notice how the Mycenaeans sit right next to the Sardinians while they should be shifted way east of that, since apart from their Anatolian Neolithic base they also score significant chg/Iran neo and some Steppe admixture.
You can confirm this by checking the 2017 Lazaridis paper in which they had done Fst for the Mycenaean and Anatolian Bronze age samples, both of which were closer to other groups compared to Sardinians (and of course would cluster closer to south Italians and Cypriots respectively on any decent pca). Do you really think Mycenaeans are closer to North Italians than to south Italians? Because that’s what that f3 outgroup stat is showing. I am talking here irrespective of whether or not we think these groups are ancestral to one another, just overall similarity.

By the same token Log4 been closer to Scotts and Lithuanians, is obviously false by any metric. They are basically a 2 way mix of Anatolian Neolithic like populations and steppe populations (around 35/44 Yamnaya) in similar proportions to modern south balkanites but very different to modern Scotts and Lithuanians, both of which cluster with northern Europeans because they have a lot more WHG compared to these samples. I would post a pca and some models but I don’t think I can post pictures.

This again doesn’t mean that modern people are unadmixed descendants off these samples but in terms of components theese Helladic MBA northern Greeks are more similar to modern populations from that area and less so to northern Europeans. Having said that and leaving that silly model for moderns in the paper aside, it's possible that these samples are a bit more informative for areas north of the Peloponnese for later periods compared to the Mycenaean samples, even though subsequent populations will certainly not be identical to them. For example later Thessalians might be a bit more derived from them but not identical to them.

My answer was directed to Kingjohn’s question on why these samples are closer to southern Balkan populations on k12b while on the paper they cluster with Scotts. They obviously don’t and and i doubt you disagree with that.
 
Last edited:
iluvatar said:
Obviously, but I never claimed that modern south balkanites are unadmixed descendants of these samples. Clearly later movements were involved as well as absorption of natives, evident in the fact that these samples score very little CHG/IRAN Neo while both Minoans and Cycladic/Helladic EBA do.

A thing to note is that their pca is way off. Notice how the Mycenaeans sit right next to the Sardinians while they should be shifted way east of that, since apart from their Anatolian Neolithic base they also score significant chg/Iran neo and some Steppe admixture.
You can confirm this by checking the 2017 Lazaridis paper in which they had done Fst for the Mycenaean and Anatolian Bronze age samples, both of which were closer to other groups compared to Sardinians (and of course would cluster closer to south Italians and Cypriots respectively on any decent pca). Do you really think Mycenaeans are closer to North Italians than to south Italians? Because that’s what that f3 outgroup stat is showing. I am talking here irrespective of whether or not we think these groups are ancestral to one another, just overall similarity.

By the same token Log4 been closer to Scotts and Lithuanians, is obviously false by any metric. They are basically a 2 way mix of Anatolian Neolithic like populations and steppe populations (around 35/44 Yamnaya) in similar proportions to modern south balkanites but very different to modern Scotts and Lithuanians, both of which cluster with northern Europeans because they have a lot more WHG compared to these samples. I would post a pca and some models but I don’t think I can post pictures.

This again doesn’t mean that modern people are unadmixed descendants off these samples but in terms of components they are more similar to them and less so to northern Europeans. Having said that and leaving that silly model for moderns in the paper aside, its possible that these samples are a bit more informative for areas north of the Peloponnese for later periods compared to the Mycenaeans samples, even though subsequent populations will certainly not be identical to them. For example later Thessalians might be a bit more derived from them but not identical to them.

My answer was directed to Kingjohn’s question on why these samples are closer to southern Balkan populations on k12b while on the paper they cluster with Scotts. They obviously don’t and and i doubt you disagree with that.
Click to expand...

Do you think people from Central Macedonia would also be good candidates for people derived from such a source? IIRC, they cluster in the same continuum as Thessalians. I could be wrong though.

Also think your analysis is spot on.
 
Do you think people from Central Macedonia would also be good candidates for people derived from such a source? IIRC, they cluster in the same continuum as Thessalians. I could be wrong though.

Also think your analysis is spot on.
Click to expand...

Your guess is as good as mine really. 2 samples from the early bronze age are way too few to draw any safe conclusions from unfortunately, and they may not be relevant to later populations but instead represent some relic of a people present in the area at that time e.g. Luwians.

As for Macedonians and Thessalians, I would expect these pops and perhaps neighboring populations to be broadly similar, but that is mostly based on intuition and some anthropology work of the previous century, that is of course often hit and miss. These issues remain to be solved by more extensive ancient dna sampling from all over the balkans.
 
real expert said:
Angela, you got wrong what my point was, probably due to the fact, that you reply to several people. Firstly, I know that you showed me the same guy, Raoul Bova who was shown on this forum several times. I took his tanned complexion to make clear that "dark" in Greek context means such a skin tone and not SSA dark. Plus, I wanted to point out, if the prediction for the two samples is accurate and not flawed, that would mean that these tested Aegeans were darker than this predynastic Egyptian who was clearly not dark brown till black but moderately light skinned. So it‘s beyond me why you think I‘m tr*lling and accuse me of not being interested on your take on this study. If that wasn't the case I wouldn't ask and discuss with you. Again, I was referring to the prediction which we all find surprising because nobody expected very dark to black ancient Greeks as this study proposes. However, we know from history and Greek sources that Ancient Greeks considered Egyptians darker than them. Besides, I talked about the steppe admixed sample since her phenotype was predicted.
Click to expand...

Let's back track a bit.

We have a portrait found in the tomb of the elite "Griffin Man" Mycenaean Warrior. We can see he's rather dark; however, there's no way of telling whether that's a tan or his actual complexion, i.e. did he have a Raoul Bova type "winter" complexion. I tend to doubt he was that light given how long ago it was and that selection for skin color has continued, but I really don't know.

We also have representations of the Egyptians of themselves in relation to "Kushites", whom they depict as black, themselves, whom they depict as dark brown, and people from the Middle East, who look rather olive skinned to me. So, I think it's highly unlikely that Bronze Age Greeks were as dark as Egyptians. One should never take the color of mummies to mean anything at all, even if you're sure of the provenance of the picture. All sorts of totally bogus pictures are floating around on the internet.

In a way, however, that's a bit irrelevant in this context. The real question is first, whether there was an error in the calculations. Second, what does "dark" or "very dark" mean in terms of Hirisplex 6.

I referred to it upthread, but didn't provide the link. I apologize. Here it is below. Go to page 17 for the skin samples.

https://hirisplex.erasmusmc.nl/pdf/hirisplex.erasmusmc.nl.pdf

Given that the sample from Crete is SLC24A5 homogeneous, and SLC45A2 heterogeneous, it's difficult for me to believe that this sample was in the darkest category.

However, in my experience Hirisplex works, so I'm befuddled. :)
 
real expert said:
One would think that researchers would try their best to avoid flawed methodology and misleading conclusions. Whatever.
After analysing this Aegean paper do you think their phenotype prediction is absolutely legit and truth worthy or do you have some reservations?
Click to expand...


do you know the genotype of the samples? are they homozygous for both alleles?
 
Archetype0ne said:
Would love your analysis of Log2.
Click to expand...

The difference between Log 2 and Log 4 is that Log 2 has less steppe. That's the long and the short of it. I'm sure you've seen all the amateur calculator results we've all posted.

I'm hesitant to accept them at face value, however.

First of all, as to the Eurogenes ones he has specifically called the accuracy of his calculators into question, especially if it involves modern populations. Perhaps we should take him at his word.

While the Dodecad ones are, imo, far more accurate, especially for Southern Europeans, I think the Globe 13 is probably more accurate than K12b.

Even so, ADMIXTURE is no substitute imo for more advanced statistical tools based on whole genomes.

One reason I'm leery of these results is the fact that some of these tools come up with people like Macedonians and Bulgarians as closest to LOG 4, for example. LOG 4 is almost 50% steppe. You only reach 50% steppe in places like England or Germany or perhaps Poland, not in the Balkans, and not, certainly, in places like Macedonia. Of course, the distance is great, 8, if I remember correctly, so that might be the reason, but still...

As for Log 2, given the amount of steppe, various places in southern Europe should be relatively close, so Tuscans aren't a terrible result, but in actuality I think Northern Italians are more likely to have that amount of steppe than Tuscans.

Albanians being, in a sense, eastern shifted Tuscans, and Thessaly being eastern and slightly south of Tuscany, we can see how they come into the picture, but I'm not sure it's correct.
 
Angela said:
We have a portrait found in the tomb of the elite "Griffin Man" Mycenaean Warrior. We can see he's rather dark; however, there's no way of telling whether that's a tan or his actual complexion
Click to expand...

Which portrait is that?
 
Status
Not open for further replies.
Back
Top