Ancient Rome: A genetic crossroads of Europe and the Mediterranean

**XenophobicPrussian** · 11-13-2019, 06:49 AM

Originally Posted by Samnium

You can obtain very good distances with a 6-population model, I've made one that works very well in central and southern italian regions.

Envoyé de mon ALE-L21 en utilisant Tapatalk

It's not about the distances, it's about missing crucial population signals that you might miss on a limited amount of source populations. I actually agree that 6 will be enough for most populations, but there are people that think even 6 is overfitting and don't go over 3-4. 3-4 will be extremely inaccurate.

I'll give you an example, say you model Italians with Celts, Germanics, Mycenaeans, Republican Romans, Natufians, and Iran_N. That's already 6. A crucial Anatolian signal will be missed(and possibly vice versa if you replaced one of them with Anatolian), Mycenaean will pick up a lot of the Anatolian admixture and total Middle-Eastern admixture isn't known. Crucial information missing. Putting multiple of each population you're trying to detect(ie multiple Celtic proxies, using both Hallstatt and Lech_MBA for example) is also useful because various groups in the G25 coordinates have outliers included in them or are really small sample sizes(Hallstatt is only 2 individuals for example, and both individuals are very different from eachother), and even more useful when using individuals/small averaged groups rather than averaged groups because you can miss out on crucial specific genetic drift that one individual may have but another may not.

**Samnium** · 11-13-2019, 07:03 AM

Originally Posted by XenophobicPrussian

It's not about the distances, it's about missing crucial population signals that you might miss on a limited amount of source populations. I actually agree that 6 will be enough for most populations, but there are people that think even 6 is overfitting and don't go over 3-4. 3-4 will be extremely inaccurate.

I'll give you an example, say you model Italians with Celts, Germanics, Mycenaeans, Republican Romans, Natufians, and Iran_N. That's already 6. A crucial Anatolian signal will be missed(and possibly vice versa if you replaced one of them with Anatolian), Mycenaean will pick up a lot of the Anatolian admixture and total Middle-Eastern admixture isn't known. Crucial information missing. Putting multiple of each population you're trying to detect(ie multiple Celtic proxies, using both Hallstatt and Lech_MBA for example) is also useful because various groups in the G25 coordinates have outliers included in them or are really small sample sizes(Hallstatt is only 2 individuals for example, and both individuals are very different from eachother), and even more useful when using individuals/small averaged groups rather than averaged groups because you can miss out on crucial specific genetic drift that one individual may have but another may not.

I think that 6 is a good number yes, 2-3 leads to enormous errors.

Well I understand that some want to breakdown all the ancient components, personally I work more by using less populations, being conscient that some populations would include other admixture. It gives a better historical explanation/model to understand the changes that took place in the past.

The other way to work is better for very ancient samples or new individuals from previously unsampled populations.

**Calpurnius** · 11-13-2019, 03:37 PM

Out of boredom I did another PCA this time only focusing on ancient and modern European samples, it's amazing how ÜBER WHG these new mesolithic Italians are, you can almost see how much of rest of HGs almost curve to the right in the direction of Anatolian-HG heavy populations as well as of course towards eastern HGs. We've found our new champions it seems.

**Creoda** · 11-13-2019, 09:16 PM

The new Mesolithic samples have put me over 50% Steppe derived

Target: Creoda_scaled
Distance: 5.8449% / 0.05844946
Aggregated
50.6 Yamnaya_RUS_Samara
36.8 Anatolia_Barcin_N
12.6 ITA_Grotta_Continenza_Meso

**Samnium** · 11-13-2019, 09:18 PM

Originally Posted by Creoda

The new Mesolithic samples have put me over 50% Steppe derived

Target: Creoda_scaled
Distance: 5.8449% / 0.05844946
Aggregated
50.6Yamnaya_RUS_Samara
36.8Anatolia_Barcin_N
12.6ITA_Grotta_Continenza_Meso

The mesolithic samples are incredible really...

They are so "pure" and distinct from other previous WHG samples.

We talk a lot about the Imperial or Iron age samples but these are as important as the other discoveries.

Envoyé de mon ALE-L21 en utilisant Tapatalk

**WeirdLookingFellow** · 11-13-2019, 09:29 PM

Originally Posted by Calpurnius

Out of boredom I did another PCA this time only focusing on ancient and modern European samples, it's amazing how ÜBER WHG these new mesolithic Italians are, you can almost see how much of rest of HGs almost curve to the right in the direction of Anatolian-HG heavy populations as well as of course towards eastern HGs. We've found our new champions it seems.

Rofl at Romanians and Bulgarians looking like they are struggling to keep Greeks and Western Europeans together