Autosomal map : African admixture (from Dodecad)

[Knovas]

Give it up, nothing is clear for him.

He uses the most insignificant details to keep the reason going against all things widely accepted. Just wish good luck for him in such desperate crusade.

[Knovas]

Going back to the thread, I noticed there's an interesting sample from the Azores isles: DOD816

It shows slightly higher East African compared to the Portuguese, but the rest of African ancestry is even lower. What it's really high in comparison is the West Asian and Southwest Asian admixture. Don't know too much history about this place, I supose population could be quite hetergeneous there.

[Wilhelm]

Going back to the thread, I noticed there's an interesting sample from the Azores isles: DOD816

It shows slightly higher East African compared to the Portuguese, but the rest of African ancestry is even lower. What it's really high in comparison is the West Asian and Southwest Asian admixture. Don't know too much history about this place, I supose population could be quite hetergeneous there.

Interesting. The high West-Asian and Southwest is possibly because of all the Jewish settlers in Azores,

[Knovas]

I didn't know there was significant Jewish presence there. The explanation makes sense, because both admixtures are present in the individual at substantial levels.

[Cambrius (The Red)]

The Azores had Jewish migration - mainly people escaping persecution from mainland Portugal and Spain. Both the Azores and Madeira are treated as genetically separate from continental Portugal since their histories are quite different.

[Drac II]

Medical studies have concluded that genetic traits found primarily in Africa (>80%) can be found at up to 20% of the Sardinian population and that there is a cline of African genetic input decreasing as one moves away from Iberia in a north easterly direction [3].

1. Sanchez-Velasco P, Gomez-Casado E, Martinez-Laso J, et al. (May 2003). "HLA alleles in isolated populations from North Spain: origin of the Basques and the ancient Iberians"

2. Choukri F, Chakib A, Himmich H, Raissi H, Caillat-Zucman S (June 2002). "HLA class I polymorphism in a Moroccan population from Casablanca". European Journal of Immunogenetics 29 (3): 205–11.

3. Gómez-Casado E, del Moral P, Martínez-Laso J, et al. (March 2000). "HLA genes in Arabic-speaking Moroccans: close relatedness to Berbers and Iberians". Tissue Antigens 55 (3): 239–49.

Thus, there is medical evidence that suggests Sardinia and Iberia share significantly higher levels of gene flow from Africa, perhaps as high as 20% in some areas. However, the relative isolation and European genetic influx into these areas has decreased the total African admixture to some degree.

HLA genes? Is this a joke? According to these same hardly reliable markers that you are trying to use for your strange little agenda against Spaniards, Greeks, unlike other Europeans from the Mediterranean area, are related to sub-Saharan Africans like Ethiopians:

"Greeks seem to share genetic HLA features (Chr 6) with Sub-Saharans. The relatedness of Greeks to Sub-Saharans has been confirmed by other studies based on chromosome 7 genetic markers." (Hajjej et al. 2011. "HLA class I and class II polymorphisms in Tunisian Berbers")

There is a reason why the majority of population geneticists have shun HLA genes: they can be very misleading. This has been known since a long time ago:

"Along this slippery way, also the work of A. Mourant et al: "The distribution of human blood groups and other polymorphisms", University of Oxford, 1976, pp 1055, is missing from their bibliography. In this edition it is referred that the HLA gene is an unreliable basis to exact conclusions on such matters, since it makes the inhabitants of Basque, Iceland and Congo to have … bonds of kinship." (Aris N. Poulianos, "The Origins of the Greeks")

[Dorianfinder]

I have done some studies using 'byseg' and 'progress' modes to view the convergence of different admixture elements within Northwest Africans and Southwest Asians. I am open to discussion but will not answer to rude replies. What I found appeared to suggest the following:

NW African elements seem to converge with West Asian and Mediterranean elements with traces of Neo African (West Africa) being a unique trait of the NW African admixture signature.

In Northeast African populations there appears to be significant Southwest Asian, which as an admixture converges in progressive iterations with West Asian and Mediterranean respectively but also includes some convergence with East African admixture components. Interestingly, these East African components together with Southwest Asian elements appear to converge with Mediterranean in the North and minute traces of Palaeo-African in the South. Whereas Northwest African appears to converge with West European and West Asian elements in the north and traces of Neo African elements in the south.

This is interesting as it differentiates Northeast Africa as having traces of Palaeo-African ancestry from the Sub-Saharan hunter-gatherer populations such as the Khoisan bushmen whereas Northwest Africa shows traces of Neo African and West African ancestry from the Sub-Saharan Negroid populations such as the Herero and Ovambo found in Angola who have approx. 100% Neo African admixture.

Just a thought: There may be Neo African elements from back-migrations that left West Europe/Mediterranean for West African many moons ago. This could potentially explain some of the high West European admixture scores in West African individuals. On second thought, the desert slave route ran through this area too. Finally it is worth mentioning the relatedness of mtdna L1, L2 and L3 and the elevated frequency of mtdna L in the Iberian peninsula and Sub-Saharan Southwest Africa.

[Knovas]

There's no such elevated MtDNA L in Iberia. The most elevated figures were got in a study in Alcacer do Sal (Portugal) including only 50 samples, which is pure joke. 0 evidence to make a claim like this.

The Eupedia "others" category say 7% for Spain, and icludes the following: The "Other" category includes mostly the older haplogroups N, R, pre-HV and HV, but also occasionally a few African (L) or Asian haplogroups (A, B, C, D, M, Z).

It worth to add that the Mediterranean component found in Iberia is mostly Southwestern, being specially dominant in the Northeast side sometimes, which is the most removed from Sub-Saharan Africans as the Euro7 Calculator shows in the Fst distances: https://docs.google.com/spreadsheet/...hl=en_US#gid=1

In short: I don't see a significant connection for remarking this. It is obviously not important, without denying some minor admixture could exist.

[Dorianfinder]

There's no such elevated MtDNA L in Iberia. The most elevated figures were got in a study in Alcacer do Sal (Portugal) including only 50 samples, which is pure joke. 0 evidence to make a claim like this.

The Eupedia "others" category say 7% for Spain, and icludes the following: The "Other" category includes mostly the older haplogroups N, R, pre-HV and HV, but also occasionally a few African (L) or Asian haplogroups (A, B, C, D, M, Z).

It worth to add that the Mediterranean component found in Iberia is mostly Southwestern, being specially dominant in the Northeast side sometimes, which is the most removed from Sub-Saharan Africans as the Euro7 Calculator shows in the Fst distances: https://docs.google.com/spreadsheet/...hl=en_US#gid=1

In short: I don't see a significant connection for remarking this. It is obviously not important, without denying some minor admixture could exist.

I was thinking more in the range of 2-3% mtdna L in Iberia with about double that on the West coast, but I would not get too excited if higher frequencies were reported as this may be due to isolated cases of sampling bias as you correctly mention. Elevated frequencies, again no more than 3%, were found in specific parts of Italy (Latium, Basilicata and Sicily) that give us a further indication of Sub-Saharan gene-flow to Southern Europe.

Concerning the Euro7 Fst distances, the African component does not incorporate all Sub-Saharan admixture elements. I am sure the Northwest African went to Southwestern. Sub-Saharan in general refers to the Capoid Palaeo-African and Negroid Neo-African elements. Northwest African admixture is a complex mix of Neo-African, Mediterranean, West Asian/Caucasus and IMHO a bit of East African originating quite possibly from the Arab sub-structure within Berber ancestry.

[Knovas]

North Africans were more Southeastern than Southwestern as I showed in the Euro7 post, and according to what you write: NW African elements seem to converge with West Asian and Mediterranean elements with traces of Neo African (West Africa) being a unique trait of the NW African admixture signature.

That means the Southeastern is the one including Northwest African and also Southwest Asian (see Turks, Armenians, etc.). Southeastern is the closest one to Caucasian (West Asian) and the closest one to Africans (Sub-Saharan), so following your argument the match is there, not in the Southwestern.

Also, the Southeastern I got (4.91%) more or less corresponded to the non European I usually get, from 0 to 10% depending on the level of resolution. However, it's the worst defined component, and it must be interpreted carefully. For example, it's obvious that the Armenian Southeastern includes more non European element than in Greeks, but it's difficult to determine the exact percent.

In a K=12 style including the Northwest African component, we'll see if it reduces a lot the Iberian Southwestern if the incoming v4 is based on this. I personally don't expect nothing drastic considering the mentioned points.

[Dorianfinder]

North Africans were more Southeastern than Southwestern

NW and NE African admixture differ as to the TYPE of African components they carry, Southeastern has traces of Neo-African elements found in what we have traditionally come to understand as NW African (without the West Asian/Caucasian and Med. components). On the flipside, Northeast African and East African have more Southwest Asian but East African has traces of Palaeo-African elements which is NOT the same as the Neo-African found in NW Africa.

Neo-African is primary Negroid (Ovambo, Herero, Bantu etc) whereas Palaeo-African is Sanid (Khoisan). More importantly though, the Palaeo-African seems to converge with the Afrasian elements unique to South Asia, this is interesting and warrants further investigation I should think. I always thought the bushmen resembled East Asians rather than Negroid populations.

This is also interesting in that the Neo-West African component (Negroid), has less SW Asian and more Central African elements (eg. Senegalese/Congolese) in general when compared to East African and SW Asian populations with Palaeo-African admixture.

This means that the African in euro7 is Palaeo-African and East African NOT the Neo-African found in the Southwestern component of the euro7 that is better known as Northwest African (without non-African components). This difference is significant as it gives us an indication of the time of introduction and the route followed from Southwest Africa to Northeast and Northwest Africa. The further one travels to the East in Northern Africa the less Neo-African one encounters! The African one finds in the Northeast Africans is more ancient than the Neo-African in the Northwest of Africa.

Hopefully the v4 calculator will provide us with better resolution wrt Palaeo-African and the ancient traces of Palaeo-African found throughout the Middle-East (1-5% in parts of Italy/Greece/Palestine/Jordan/Yemen etc.) that appear to have a very different story to that of the Sanid hunter-gatherers of Southern Africa. To resolve this Pygmy and San control groups would have to be included in a Middle-Eastern and Eastern Mediterranean population group, with primary components being Mediterranean, SW Asian, W Asian, Neo African, East African and Palaeo-African.

Similarly, it would be great to see some more resolution of the Northwest African into 'purely' Afrasian and 'newly' introduced Eurasian elements.

[Knovas]

What you say makes no sense. The Palaeo African is the most Negroid component clearly attested by the v3 distances. The most removed from Europe, and not closer to West Asian, Southwest Asian and South Asian than the Neo African is, since this one is REALLY the component which deviates a little towards Eurasia, although it's almost entirely Negroid. Even comparing with Northeast and Southeast Asian, Neo African it's still closer...impossible to sostain such thing about Palaeo African.

100% sure: no convergence between Palaeo African and Eurasia. Please, re-check again: http://dodecad.blogspot.com/search?u...max-results=12

It's a true Negroid component with independence of the populations having such admixture. FACT.

The Southeastern is clearly the component absorbing the mentioned elements, since it's the worst defined between all; Dienekes' must fix it in next runs. And for the moment, in regards for the Southwestern component, the only clear match looking at ancient populations is haplogroup I2a1a* and similars, which has absolutely nothing to do with Africa. That explains perfectly the evident drift showed in the Fst distances while checking the component in comparison with the others.

[Cambrius (The Red)]

There's no such elevated MtDNA L in Iberia. The most elevated figures were got in a study in Alcacer do Sal (Portugal) including only 50 samples, which is pure joke. 0 evidence to make a claim like this.

The Eupedia "others" category say 7% for Spain, and icludes the following: The "Other" category includes mostly the older haplogroups N, R, pre-HV and HV, but also occasionally a few African (L) or Asian haplogroups (A, B, C, D, M, Z).

It worth to add that the Mediterranean component found in Iberia is mostly Southwestern, being specially dominant in the Northeast side sometimes, which is the most removed from Sub-Saharan Africans as the Euro7 Calculator shows in the Fst distances: https://docs.google.com/spreadsheet/...hl=en_US#gid=1

In short: I don't see a significant connection for remarking this. It is obviously not important, without denying some minor admixture could exist.

Indeed, the Alcacer do Sal study leaves much to be desired. If one understands the history of this particular village, from the late 1400s on, he would realize instantly that Sal's genetics are not representative of the Portuguese population. You see, Alcacer do Sal was a socially isolated community made up mainly of Morrish / N. African, Arab and negroid slaves / ex-slaves along with people who had been ostracized due to certain diseases (e.g., leprosy). Adding to that, the area was a malaria zone. Alcacer had little to do with surrounding communities for long periods of time. Geneticists who examined the 2005 study commented that samplings from Alcacer do Sal should have been segregated or not used at all since they do not accurately reflect indigenous Portuguese types.

[Knovas]

And I add:

The Euro7 Calculator includes Mandenka and Yoruba, characterized for their strong Neo African admixture. They become 99-100% African.

No problem to read it.

[Wilhelm]

Finally it is worth mentioning the relatedness of mtdna L1, L2 and L3 and the elevated frequency of mtdna L in the Iberian peninsula and Sub-Saharan Southwest Africa.

Elevated frequency of mtDNA L in Iberia ? In a sample of 686 spaniards there was 0% of mtDNA L (Rhouda et al. 2006)

[Vallicanus]

Elevated frequency of mtDNA L in Iberia ? In a sample of 686 spaniards there was 0% of mtDNA L (Rhouda et al. 2006)

Then how do you explain these Achilli et alia's 2007 figures for subsaharan African mtDNA:

Spanish Galicia...3.7pc
NE Spain...1.68pc
Central Spain...0.68pc
Basque Country...0.64pc

The latest Dienekes and Eurogenes surveys also show some SS African mixture in Spain.

[Knovas]

It's called noise. Most of this scores are less than 0.5% and, specially in the last Eurogenes K=10, the Sub-Saharan cluster showed some of this figures even in Russians.

For more you repeat, it won't never be relevant.

[Dorianfinder]

What you say makes no sense. The Palaeo African is the most Negroid component clearly attested by the v3 distances. The most removed from Europe, and not closer to West Asian, Southwest Asian and South Asian than the Neo African is, since this one is REALLY the component which deviates a little towards Eurasia, although it's almost entirely Negroid. Even comparing with Northeast and Southeast Asian, Neo African it's still closer...impossible to sostain such thing about Palaeo African.

100% sure: no convergence between Palaeo African and Eurasia. Please, re-check again: http://dodecad.blogspot.com/search?u...max-results=12

It's a true Negroid component with independence of the populations having such admixture. FACT.

The Southeastern is clearly the component absorbing the mentioned elements, since it's the worst defined between all; Dienekes' must fix it in next runs. And for the moment, in regards for the Southwestern component, the only clear match looking at ancient populations is haplogroup I2a1a* and similars, which has absolutely nothing to do with Africa. That explains perfectly the evident drift showed in the Fst distances while checking the component in comparison with the others.

It does make sense, you don't understand it that's all. The components used to form the components known as Palaeo-African and Neo-African do not make a difference as we are talking convergence during progressive iterations during the admixture calculator's search for best fit. Change 'verbose' in the dv3.par file to 'progress' and look at the progress of your admixture percentages. You will see Southwestern increase as African decreases to end at 0%. The same is true for Palaeo-African and Southeastern. The convergence criteria are good so no significant statistical noise to blame for the progression.

You are very defensive and quick to criticize, have you been right your entire life? It's okay to be wrong sometimes, your attitude is almost teenager like sometimes. Anyhow, you don't have to feel threatened by what other people say, it's just a discussion forum.

[Wilhelm]

Then how do you explain these Achilli et alia's 2007 figures for subsaharan African mtDNA:

Spanish Galicia...3.7pc
NE Spain...1.68pc
Central Spain...0.68pc
Basque Country...0.64pc

The latest Dienekes and Eurogenes surveys also show some SS African mixture in Spain.

WRONG. The last Eurogenes run (WE_10) shows 0% of SSA in spaniards.

[Knovas]

It does make sense, you don't understand it that's all. The components used to form the components known as Palaeo-African and Neo-African do not make a difference as we are talking convergence during progressive iterations during the admixture calculator's search for best fit. Change 'verbose' in the dv3.par file to 'progress' and look at the progress of your admixture percentages. You will see Southwestern increase as African decreases to end at 0%. The same is true for Palaeo-African and Southeastern. The convergence criteria are good so no significant statistical noise to blame for the progression.

You are very defensive and quick to criticize, have you been right your entire life? It's okay to be wrong sometimes, your attitude is almost teenager like sometimes. Anyhow, you don't have to feel threatened by what other people say, it's just a discussion forum.

What you say is simply wrong and I must tell it. I f you don't like it it's not my problem.

What you told about Palaeo African using the "Sanid" term and linking it with Eurasia (similarity with East Asians) it's plain false as attested by the Fst distances. Even a child could be aware of this looking a few seconds to the table.

It's a true negroid component, and it's Neo African the one deviating towards Eurasians just a little, again, attested by the Fst distances. So if you are not agree with this, I think you should tell Dienekes' since he is the one who calibrates the instrumental and names the clusters.

What is a teenager attitude is to deny such obvious thing. Just see the noise scores reported in Madenka, being first Northeastern (0.5), Northwestern (0.4) and finally Southwestern (0.1), all of them with peaks in mainland Europe. And curiously, Southestern with the highest peak in Armenians, tended to an absolute 0 in a mostly Neo African population when this should happen, according to you, in a mostly Palaeo African population. So no sorry, the correlation you mention it does not exist.

The Southeastern appears as an ambiguous cluster valid to absorve different elements in THIS calculator, while it cannot be said about the three others having such distances, much less considering the huge drift in the Southwestern (the best defined of ALL West Eurasian components). So as Dienekes' intruduces them again in a K=12 style, surely a "pure" form of Southeastern would appear. I think it's clearly the main thing to improve in the mentioned run, since the rest seems to work fine in general terms.

[Vallicanus]

WRONG. The last Eurogenes run (WE_10) shows 0% of SSA in spaniards.

Believe your fantasies if they help you get up in the morning.

[Knovas]

Believe your fantasies if they help you get up in the morning.

Keep thinking scores listed as less than 1% are relevant if this helps you to wake up in the morning.

Empty discussion.

[Dorianfinder]

What you say is simply wrong and I must tell it. I f you don't like it it's not my problem.

What you told about Palaeo African using the "Sanid" term and linking it with Eurasia (similarity with East Asians) it's plain false as attested by the Fst distances. Even a child could be aware of this looking a few seconds to the table.

It's a true negroid component, and it's Neo African the one deviating towards Eurasians just a little, again, attested by the Fst distances. So if you are not agree with this, I think you should tell Dienekes' since he is the one who calibrates the instrumental and names the clusters.

What is a teenager attitude is to deny such obvious thing. Just see the noise scores reported in Madenka, being first Northeastern (0.5), Northwestern (0.4) and finally Southwestern (0.1), all of them with peaks in mainland Europe. And curiously, Southestern with the highest peak in Armenians, tended to an absolute 0 in a mostly Neo African population when this should happen, according to you, in a mostly Palaeo African population. So no sorry, the correlation you mention it does not exist.

The Southeastern appears as an ambiguous cluster valid to absorve different elements in THIS calculator, while it cannot be said about the three others having such distances, much less considering the huge drift in the Southwestern (the best defined of ALL West Eurasian components). So as Dienekes' intruduces them again in a K=12 style, surely a "pure" form of Southeastern would appear. I think it's clearly the main thing to improve in the mentioned run, since the rest seems to work fine in general terms.

I will explain it a different way. First-off, Fst distances are simply distances from a manufactured mean, they should not be taken literally as admixture percentages or real value norms, the concept of a PCA (principle component analysis) is something one should understand first before being able to interpret Fst distances. More importantly though, if one understands the constructs in statistics then the norms are seen as facilitative approximations to a mean that for all intents and purposes has very little value other than that it helps to fix one's conceptualization mathematically.

The Sanid-Capoid bushmen and Sandawe are grouped as Palaeo-African. This means that they have much less Afrasian admixture than the Congolid populations classified phenotypically as Negroid. The latter have up to 80% Afrasian admixture whereas the Palaeo-African populations have less that 20%. Southwest Asian in Africa converges with a small but significant Palaeo-African component. The Sanid-Capoid bushmen are the only group in Africa to display epicantic eye-lids such as we find in East Asian populations. Similarly, Sanid-Capoid ydna haplogroups are the most diversified and are mostly HGs A and B. Interestingly, other first nation peoples from Eurasia belong to HG C1 and carry the same gene for epicantic eye-lids.

In Northeast Africa, the Middle-East and Southeast Med. small amounts of Palaeo-African are common and point to a different origin of Sub-Saharan ancestry than the Northwest African Congolid elements found primarily in West African Neo-Africans.

African admixture is more complex than Negroid FACT! Find the Fst distances for the various components that make up African admixture so we can discuss this properly using population groups and their respective admixture of East African, Neo-African and Palaeo-African.

Remember that different dodecad calculators have different constituent components for Sub-Saharan and South African.

[Knovas]

Well Dorian, I don't pretend to hurt you first of all, but I think you are complicating the argument to keep the reason when the discussion is over. The distances are perfectly clear, there's nothing more to understand in regards for the Palaeo African position.

The features that you identify as "East Asian" in the Bushmen have really NOTHING to do with Asia. Palaeo African populations have usually a curious diversity in the facial traits, but it's just due to this: they are very Paleolithic and have retained most of the "original modern human seed" to make it easy to understand. In other words, they are more related to the "homos" who evolved in different forms all over the world.

I mean, this features (doesn't matter if "Asian looking" or something else) are absolutely native to Africa, hence, associated to Negroids. It's not the same, even not slightly close in no way, to what you find among any Mongoloid population in genetic terms. They are Pre-Asian.

My English vocabulary perhaps is not enough rich to explain it in a more proper way, but I think the idea it's clear. Palaeo Africans are true negroids with independence of their strange (or not) appearence. But if you prefer to avoid the term Negroid: Palaeo Afrians are more African genetically speaking than Neo Africans.

[Wilhelm]

Believe your fantasies if they help you get up in the morning.

It's not my fantasies, you can see the percentages here, West-Eurasia K=10 (the last Eurogenes run) :

http://bga101.blogspot.com/2011/09/g...ross-west.html

Sub-Saharan for Spaniards : 0%