PDA

View Full Version : First Genome of Ancient African Suggests Eurasian Stream Into Africa



Angela
08-10-15, 20:57
It's going to be in Science Magazine:
http://www.sciencemag.org/content/350/6257/149.full.pdf

"A paper published online this week in Science (http://scim.ag/MGLlorente) reveals the first prehistoric genome From Africa: that of Mota, a hunter-gatherer man who lived 4500 years ago in the highlands of Ethiopia."

"genomes suggest a major migration into Africa by farmers from the Middle East, possibly about 3500 years ago. These farmers'DNA reached deep into the continent, spreading even to groups considered isolated, such as the Koisan of South Africa and the pygmies of the Congo."


"Population geneticist David Reich of Harvard University is struck by the magnitude of the mixing between Africans and Eurasians. He notes that “a profound migration of farmers moving from Mesopotamia to North Africa has long been speculated.” But, he says, “a western Eurasian migration into every population they study in Africa—into the Mbuti pygmies and the Khoisan? That's surprising and new.”"

Well, well.

Angela
08-10-15, 22:03
Here is the Abstract:
"Characterizing genetic diversity in Africa is a crucial step for most analyses reconstructing the evolutionary history of anatomically modern humans. However, historic migrations from Eurasia into Africa have affected many contemporary populations, confounding inferences. Here, we present a 12.5x coverage ancient genome of an Ethiopian male (‘Mota’) who lived approximately 4,500 years ago. We use this genome to demonstrate that the Eurasian backflow into Africa came from a population closely related to Early Neolithic farmers, who had colonized Europe 4,000 years earlier. The extent of this backflow was much greater than previously reported, reaching all the way to Central, West and Southern Africa, affecting even populations such as Yoruba and Mbuti, previously thought to be relatively unadmixed, who harbor 6-7% Eurasian ancestry."

Supplementary Info:
http://www.sciencemag.org/content/suppl/2015/10/07/science.aad2879.DC1/Gallego-Llorente.SM.pdf

He was mtDna L3x2, which is restricted to th Horn of Africa and the Nile Valley in modern Ethiopian samples, which to the authors suggests maternal continuity in that country over the last 4500 years.

The same goes for his y DNA E1b1.

The sample is closest to the ARI, who speak an Omotic language, the most differentiated of the Afro-Asiatic languages, and show "large West Eurasian components (17.8±1.0% and 14.9±1.2% for Ari Cultivator and Ari Blacksmith respectively)."

"D statistics and f4 ratios show that Mota has no discernible Neanderthal component", as opposed to modern day Africans. So the Neanderthal in Africans comes from West Eurasian introgression into Africa. Mota also had no Denisova component.

He had altitude adaptation snps but no de-pigmentation snps or lactase persistence snps.

This is the West Eurasian percentage in East Africans based on Yoruba/Druse:

7445

No wonder some East Africans on 23andme were only coming up as 50% African.

MOESAN
08-10-15, 22:38
Thanks, Angela - very interesting, not too surprising concerning some SSA ethnies but very amazing for others...

Angela
09-10-15, 00:26
Thanks, Angela - very interesting, not too surprising concerning some SSA ethnies but very amazing for others...

It reminds me of the situation with Mexicans. I was amazed at how almost every Mexican genome shows around 5% SSA, yet the historical record only shows a very limited importation of African slaves into specific areas. There's more of it at the site of entry, but a small bit shows up virtually everywhere.

In some ways it's also like West Eurasian in the Americas. Even tribes far in the interior of the Amazon sometimes show up with a bit of West Eurasian.

Angela
09-10-15, 01:39
OK Angela, but if Neolithic different populations in Europe show very different distributions of genes - even if these genes are all of them found in say: Anatolia - we can try to know if the causes are more recent drifts after reaching Europe OR are differences pre-existing in the source great region; even if today all these genes are evenly distributed in Anatolia (it's an example, pure theory) they could have had very different distributions too in ancient Anatolia... Just splitting hairs.
All that will not change our ffuture.
Bone net (regional french)



No, it won't. :)

If they can recover all this ancient dna from Ethiopia, they should be able, at some point, to sequence lots of farmers and hunter-gatherers from the Near East, and then we'll know how much substructure there was in different periods.

I don't know if it matters in terms of Europe, however. I suppose it depends on whether the homogenization took place in southeastern Anatolia/north-central Levant and then part of the group left for Cyprus and up into Greece, and another part of the group followed the coast of Anatolia or just headed north into central Anatolia, or the group that left for Cyprus and may have traveled by sea (Cardial?) was one group and the group that colonized Anatolia was different. I'm not sure that the latter scenario is supported, however. It's true that the Hungarian Neolithic is a little different from LBK, which may be a little different from the Spanish Early Neolithic etc. On another level, they're all pretty similar, however, I think.

Everything depends on the lens one chooses, yes? Perhaps a good analogy is what came out of the Leslie et al study of British genomes? Looking at very, very subtle differences, the authors were able to find substructure in the people of Britain. However, if you pull back a little bit, say to the level of a company like 23andme, they're one population, and in fact at the edges it's difficult to tell them apart from the Dutch and the Danes.

So, if we look at LBK, a second wave in the mid-to-late Neolithic that brought J2 and E-V13, perhaps some Bronze Age flow from West Asia to Crete and then into the Balkans and the Central Mediterranean, etc., maybe some later migrations, and of course the Indo-Europeans, and if one ignores varying amounts of WHG, ANE, and other minor components, and just focuses on the "farmer" ancestry, the discovery of differences might depend on the "lens" being used. If one looks in a "fine grained" way there are probably differences. However, on another level, it's all the same.

The paper on the Ancient African is instructive. The West Eurasian component that moved south 1500 BC, if their dating is correct, was still very Stuttgart like.*

Ed. You know, that's a bit of an issue for some theories if that's the right date. (Oh, this dating stuff!)

If the West Eurasian that came to them has no ANE, then when would the ANE have arrived in the Levant, and with whom?

LeBrok
09-10-15, 07:13
Surprising dating. I would guess that Early Neolithic farmers expended, or at least started expending, into SSA as soon as they populated Egypt, way before ancient Egyptian history starts 3,000 BC. Afterwords we could see consecutive expansions.

bicicleur
09-10-15, 08:45
Interesting to learn that haplo E didn't have any Neanderthal or Denisovan admixture.
So all Neanderthal admixture in Africa (there is, alltough less than in other parts of the world) comes from other sources.

Anatolian cattle herders entered Africa during or right after the 8.2 ka climate event, so that is when admixture started, yet this 4.5 ka Mota cave individual wasn't affected.

I guess pygmee and Khoisan tribes got admixture indirectly, through Bantu tribes.

bicicleur
09-10-15, 08:50
I wonder whether this is correct :

Then about 3,000 years ago, a group of farmers from the Middle East and present-day Turkey came back to the Horn of Africa (probably bringing crops like wheat, barley and lentils with them).

afaik Arab tribes (J1) trading and cultivating frankincense and myrhe entered Ethiopia from western Yemen some 3000 years ago (cfr the Queen of Sheba)
these Arab tribes knew all of these crops

yet that doesn't explain the DNA similarities between Ari people and LBK/Sardinians

bicicleur
09-10-15, 10:24
No, it won't. :)

If they can recover all this ancient dna from Ethiopia, they should be able, at some point, to sequence lots of farmers and hunter-gatherers from the Near East, and then we'll know how much substructure there was in different periods.

I don't know if it matters in terms of Europe, however. I suppose it depends on whether the homogenization took place in southeastern Anatolia/north-central Levant and then part of the group left for Cyprus and up into Greece, and another part of the group followed the coast of Anatolia or just headed north into central Anatolia, or the group that left for Cyprus and may have traveled by sea (Cardial?) was one group and the group that colonized Anatolia was different. I'm not sure that the latter scenario is supported, however. It's true that the Hungarian Neolithic is a little different from LBK, which may be a little different from the Spanish Early Neolithic etc. On another level, they're all pretty similar, however, I think.

Everything depends on the lens one chooses, yes? Perhaps a good analogy is what came out of the Leslie et al study of British genomes? Looking at very, very subtle differences, the authors were able to find substructure in the people of Britain. However, if you pull back a little bit, say to the level of a company like 23andme, they're one population, and in fact at the edges it's difficult to tell them apart from the Dutch and the Danes.

So, if we look at LBK, a second wave in the mid-to-late Neolithic that brought J2 and E-V13, perhaps some Bronze Age flow from West Asia to Crete and then into the Balkans and the Central Mediterranean, etc., maybe some later migrations, and of course the Indo-Europeans, and if one ignores varying amounts of WHG, ANE, and other minor components, and just focuses on the "farmer" ancestry, the discovery of differences might depend on the "lens" being used. If one looks in a "fine grained" way there are probably differences. However, on another level, it's all the same.

The paper on the Ancient African is instructive. The West Eurasian component that moved south 1500 BC, if their dating is correct, was still very Stuttgart like.*

Ed. You know, that's a bit of an issue for some theories if that's the right date. (Oh, this dating stuff!)

If the West Eurasian that came to them has no ANE, then when would the ANE have arrived in the Levant, and with whom?

the stronger the lense becomes, it becomes more and more appearant that there were far more migrations than ever suspected

now it is a point to date all these migrations and find the causes

Alan
09-10-15, 14:42
Strong argument for a back to Africa migration of Haplogroup E? As claimed in the past, since E is the closest cousin to Eurasian Haplogroup D and ED on the other hand close relatives of CF.....

E would have started of somewhere in Eurafrica (Between Arabia and East Africa), to the North F (in Mesopotamia?) to the East D and C (Iranian_Plateau, South_Central Asia and East Asia/Oceania).

Looks like the Proto SSA Haplogroups might be at the end of the day A and B? And E represents just another OoA migration.

Alan
09-10-15, 15:10
Now I see this individual was already yDNA E.

But than I don't think all back to Africa migration started 2500 BC.

Alone the Afro_Asiatic expansion in North Africa dates further back. I still suspect with older samples we will find more Eurasian admixture in Africa. And E being in reality a back migration.

Taranis
09-10-15, 15:21
The sample is closest to the ARI, who speak an Omotic language, the most differentiated of the Afro-Asiatic languages, and show "large West Eurasian components (17.8±1.0% and 14.9±1.2% for Ari Cultivator and Ari Blacksmith respectively)."

I wanted to comment on the above quote in particular. I've argued on several instances in the past that in my opinion, Afro-Asiatic languages originated in the as an early farmer's language Near East , despite the fact that only one branch of Afroasiatic (Semitic) is spoken there today. Barring that, it made sense for me to at least assume the homeland was located at the Mediterranean, since three of the major branches of Afroasiatic (Berber, Egyptian and Semitic) are located there, and you can assume that the Chadic speakers expanded too from the Mediterranean (there's also R1b-V88, common amongst the Chadic-speaking peoples and their neighbours, which I have speculated upon in the past that it basically represents an Afroasiatic branch of R1b). The only Afroasiatic branches where the Mediterranean connection is absent is with the Omotic languages and the Kushitic languages. In so far, this new genetic evidence backs up that even Proto-Kushitic and Proto-Omotic may originate from further north.

Angela
09-10-15, 15:42
I wanted to comment on the above quote in particular. I've argued on several instances in the past that in my opinion, Afro-Asiatic languages originated in the as an early farmer's language Near East , despite the fact that only one branch of Afroasiatic (Semitic) is spoken there today. Barring that, it made sense for me to at least assume the homeland was located at the Mediterranean, since three of the major branches of Afroasiatic (Berber, Egyptian and Semitic) are located there, and you can assume that the Chadic speakers expanded too from the Mediterranean (there's also R1b-V88, common amongst the Chadic-speaking peoples and their neighbours, which I have speculated upon in the past that it basically represents an Afroasiatic branch of R1b). The only Afroasiatic branches where the Mediterranean connection is absent is with the Omotic languages and the Kushitic languages. In so far, this new genetic evidence backs up that even Proto-Kushitic and Proto-Omotic may originate from further north.

Yes, I think the genetics here clearly backs up the linguistic analysis.

Angela
09-10-15, 16:22
@Alan. I agree that this has nothing to do with yDna "E". Mota is E1b1 after all. If "E" arose in Eurasia and back migrated into Africa then it happened much earlier in time...perhaps the Paleolithic, and the "original" African y lines would be A and B.

@Bicicleur,
This paper nicely correlates with the prior Pickrell paper. This component would have slowly diffused south and west, eventually even making it into groups like the San and Mbuti and South Africans, albeit at much smaller levels.

The dating certainly presents some challenges. This "Sardinian like" or "Stuttgart like" component obviously didn't reach this area until after 2500 BC, because the dating of Mota seems pretty certain. Also, as you say, there is evidence for cattle herders reaching Africa pretty early on. As LeBrok mentioned, there must have been a flow into Egypt at least by 3000 BC.

So, is this rather recent date just a function of the isolation of this particular area, and the movement south began much earlier? That would help explain why there's no ANE.

However, what would then be amazing is that these people, if they started migrating south as early as 6,000 or 5,000 BC, were still "Stuttgart like" after 4,000 years in transit. Maybe there's a parallel with what happened in Europe, but even more extreme? If the Lazaridis paper is saying that the Near Eastern farmers picked up about 10% WHG upon their initial move into Europe (maybe around the Danube gates?), then in the next three thousand years they only picked up about another 10-15 points of WHG.

That begs the question as to why, in Africa, they suddenly started admixing.

I also think there had to eventually be some gene flow going the other way, at least in places like Egypt. The Nile was an easy route for some of these perhaps admixed people to flow back north. We have a Pharaoh with a very SSA y line after all.

Ed. Oh, any guesses as to the yDna lines involved? How about T in addition to V88?

bicicleur
09-10-15, 17:27
I wanted to comment on the above quote in particular. I've argued on several instances in the past that in my opinion, Afro-Asiatic languages originated in the as an early farmer's language Near East , despite the fact that only one branch of Afroasiatic (Semitic) is spoken there today. Barring that, it made sense for me to at least assume the homeland was located at the Mediterranean, since three of the major branches of Afroasiatic (Berber, Egyptian and Semitic) are located there, and you can assume that the Chadic speakers expanded too from the Mediterranean (there's also R1b-V88, common amongst the Chadic-speaking peoples and their neighbours, which I have speculated upon in the past that it basically represents an Afroasiatic branch of R1b). The only Afroasiatic branches where the Mediterranean connection is absent is with the Omotic languages and the Kushitic languages. In so far, this new genetic evidence backs up that even Proto-Kushitic and Proto-Omotic may originate from further north.

IMO opinion, this is the origin of Afro-Asiatic : https://en.wikipedia.org/wiki/Qadan_culture

they were HG with some incipient agriculture (like Natufians)

IMO E1b1b

IMO Semitic arrived in the Levant only 4th Mill BC, with desertification of the Sahara

https://en.wikipedia.org/wiki/Proto-Semitic_language

bicicleur
09-10-15, 17:48
Strong argument for a back to Africa migration of Haplogroup E? As claimed in the past, since E is the closest cousin to Eurasian Haplogroup D and ED on the other hand close relatives of CF.....

E would have started of somewhere in Eurafrica (Between Arabia and East Africa), to the North F (in Mesopotamia?) to the East D and C (Iranian_Plateau, South_Central Asia and East Asia/Oceania).

Looks like the Proto SSA Haplogroups might be at the end of the day A and B? And E represents just another OoA migration.

if you have the time read this

http://www.jstor.org/stable/10.1086/657397

Nubyan complex : people from Africa arrived in Dhofar > 106 ka and spread all over Arabia
http://www.aggsbach.de/2013/05/the-nubian-compley/
compare with age estimates by YFull : CT wasn't born yet, this was BT
Nubyan complex was in Egypt 118 +/- 7 ka but dissapeared to reappear later, but dating lacks.
http://www.aggsbach.de/2013/08/the-early-nubian-complex/
This reappearance was backmigration of B, followed by backmigration of E.
D moved east till Sundaland.
CT persisted in Arabia roaming the desert in small bands till at least 55 ka :
https://forwhattheywereweare.wordpress.com/?s=wadi+surdud
Then both C and F expanded +/- 50 ka from Arabia, but before that both got Neanderthal admixture

It looks like during LGM E1b1b-M215 was centered around the Ethiopian Highlands which caught monsoon rains while the Sahara had expanded far south.
http://cdn.eupedia.com/images/content/E1b1b-tree.gif

arvistro
09-10-15, 21:52
I can't contribute much to this thread, just express my joy that also Africa starts to get adna coming.

MOESAN
09-10-15, 23:45
No, it won't. :)

If they can recover all this ancient dna from Ethiopia, they should be able, at some point, to sequence lots of farmers and hunter-gatherers from the Near East, and then we'll know how much substructure there was in different periods.

I don't know if it matters in terms of Europe, however. I suppose it depends on whether the homogenization took place in southeastern Anatolia/north-central Levant and then part of the group left for Cyprus and up into Greece, and another part of the group followed the coast of Anatolia or just headed north into central Anatolia, or the group that left for Cyprus and may have traveled by sea (Cardial?) was one group and the group that colonized Anatolia was different. I'm not sure that the latter scenario is supported, however. It's true that the Hungarian Neolithic is a little different from LBK, which may be a little different from the Spanish Early Neolithic etc. On another level, they're all pretty similar, however, I think.

Everything depends on the lens one chooses, yes? Perhaps a good analogy is what came out of the Leslie et al study of British genomes? Looking at very, very subtle differences, the authors were able to find substructure in the people of Britain. However, if you pull back a little bit, say to the level of a company like 23andme, they're one population, and in fact at the edges it's difficult to tell them apart from the Dutch and the Danes.

So, if we look at LBK, a second wave in the mid-to-late Neolithic that brought J2 and E-V13, perhaps some Bronze Age flow from West Asia to Crete and then into the Balkans and the Central Mediterranean, etc., maybe some later migrations, and of course the Indo-Europeans, and if one ignores varying amounts of WHG, ANE, and other minor components, and just focuses on the "farmer" ancestry, the discovery of differences might depend on the "lens" being used. If one looks in a "fine grained" way there are probably differences. However, on another level, it's all the same.

The paper on the Ancient African is instructive. The West Eurasian component that moved south 1500 BC, if their dating is correct, was still very Stuttgart like.*

Ed. You know, that's a bit of an issue for some theories if that's the right date. (Oh, this dating stuff!)

If the West Eurasian that came to them has no ANE, then when would the ANE have arrived in the Levant, and with whom?

OK Angela, I know you have a solid brain: my lens resolution (english?) concerned History: not a final result but ancient results; concerning UK, some small clusters are for a part the result of enough recent events by isolation (LATER); isolation could have played too in ancient Anatolia, but BEFORE INTRODUCTION INTO eUROPE; and theorically speaking concerning history, there is a difference between differenciation of genomes by new mutations in big groups separated for a long time and differenciation of genomes by lost of already existing genes by isolation of an already well mixed group but limited in size; an acute scrutiny of genome could I suppose tell one case from the other?
in fine, all that could seem futility? I don't know...

Tomenable
10-10-15, 23:59
affecting even populations such as Yoruba (...) previously thought to be relatively unadmixed, who harbor 6-7% Eurasian ancestry.
Razib Khan noticed ENF ("Sardinian-like") admixture in Yoruba back in 2012!:

http://blogs.discovermagazine.com/gnxp/2012/03/we-are-all-sardinians/#.VhmJzUoyU8E

The_Lyonnist
05-12-15, 14:21
Eurasians are the ancestors of Africans.

oriental
05-12-15, 22:47
Don't forget the Milankovitch cycles for climate change.

Every 20,000 years the Sahara and the Middle east turned green for 5,000 years. The last time the Sahatra and Mid East was green 5,000 years ago i.e. 3,000 BC. From 8,000 BC to 3,000 BC Sahara and Mid East were green. Before that 28,000 BC to 23,000 BC Sahara and Mid East were green and before that 48,000 BC to 43,000 BC then 68,000 BC to 63,000 BC. This is the series.

Tomenable
05-12-15, 23:02
Eurasians are the ancestors of Africans.

But only of one particular (and rather small) group of Africans, not of all or even most of Africans.

Angela
29-01-16, 18:30
There's been a correction to the Mota paper:
http://www.nature.com/news/error-found-in-study-of-first-ancient-african-genome-1.19258


"Pontus Skoglund, a population geneticist at Harvard Medical School in Boston, Massachusetts, says that he was surprised by the claim that as much as 6–7% of the ancestry of West and Central African groups came from the Eurasian migrants. But after obtaining the Mota man’s genome from Manica’s team, he and his colleague David Reich carried out their own comparison and found no evidence for that conclusion. They informed Manica’s team, who then discovered the processing error.


“Almost all of us agree there was some back-to-Africa gene flow, and it was a pretty big migration into East Africa,” says Skoglund. “But it did not reach West and Central Africa, at least not in a detectable way.” The error also undermines the paper’s original conclusion that many Africans carry Neanderthal DNA (inherited from Eurasians whose ancestors had interbred with the group)."

If Mota shows Eurasian influence, then that will affect the percentages of Eurasian ancestry in East African populations that can be held to have derived from post-Mota migrations.

Tomenable
29-01-16, 19:00
Thanks Angela,

It seems that errors happen in such studies, not just in autosomal but also uniparental stuff.

For example Mathieson et al. wrongly described ESP14 as R1b, while in fact it was R1a-M198*:

http://www.eupedia.com/forum/threads/31924-Thuringia-just-correlation-or-causation-as-well?p=474921&viewfull=1#post474921

Smal from Anthrogenica says that it was not R1b. According to Genetiker it was R1a-M198*.

I think they are both right in this case.

RISE1 - also described as R1b in one paper - is too low coverage to say anything other than R or R1.

He is a similar case as Iberian ATP3, for which Genetiker claims R1b-M269, but it is very low coverage.

Tomenable
29-01-16, 19:03
Pontus Skoglund, a population geneticist at Harvard Medical School in Boston, Massachusetts, says that he was surprised by the claim that as much as 6–7% of the ancestry of West and Central African groups came from the Eurasian migrants. But after obtaining the Mota man’s genome from Manica’s team, he and his colleague David Reich carried out their own comparison and found no evidence for that conclusion. They informed Manica’s team, who then discovered the processing error.

This new conclusion means that agriculture was invented in Central Africa independently, rather than spreading there from Eurasia.

By some Proto-Bantu ancestors of the Bantu, perhaps (?).

Or - alternatively - agriculture could get there from north of the Sahara, but through a cultural transition, without any gene flow.

Athiudisc
29-01-16, 19:10
The error also undermines the paper’s original conclusion that many Africans carry Neanderthal DNA (inherited from Eurasians whose ancestors had interbred with the group)."

I've been very curious about this lately. When Neandertal admixture was confirmed as likely for non-Africans, it was thought that the vast majority of SSAs had none. Now we're seeing some low Neandertal ancestry is some eastern African groups, but why didn't we find this earlier? Why was the conclusion over the past several years that SSAs didn't have any?

I'm sure there's a simple answer. Lack of coverage?

Angela
29-01-16, 19:11
Here is the link to the actual erratum:
https://dl.dropboxusercontent.com/u/26978112/Erratum%20with%20figures.pdf

This probably means we're back to the conclusion that there's no Neanderthal in West Africans.

It also means it's a good idea, in my opinion, to use SSA based on West Africans when trying to estimate "African" gene flow.

@Tomenable:
There's nothing to indicate that farming was invented independently in Central Africa. I don't think this affects the gene flow down from Eastern Africa, although it might change percentages slightly.

Tomenable
29-01-16, 19:37
There's nothing to indicate that farming was invented independently in Central Africa.

Really? It seems that they domesticated totally different plants / crops than farmers up north:

http://apworldipedia.com/index.php?title=Key_Concept_1.2_The_Neolithic_Revo lution_and_Early_Agricultural_Societies

https://commons.wikimedia.org/wiki/File:Centres_of_origin_and_spread_of_agriculture_v 2.svg

http://www.oxfordbibliographies.com/view/document/obo-9780199846733/obo-9780199846733-0115.xml

https://s-media-cache-ak0.pinimg.com/736x/29/f7/76/29f7765ffa61b1529d0d4dbcb8c21c82.jpg

Tomenable
29-01-16, 19:42
It remains unclear whether Sub-Saharan agriculture developed independently or was imported. Genetic evidence may be conclusive in determining what happened, and if there was no or just very little gene flow, then apparently they developed it independently:

http://www.fao.org/wairdocs/ilri/x5462e/x5462e0e.htm


The origins of agriculture in sub-Saharan Africa remain unclear. Vegeculture was probably practiced on the northern margins of the tropical forest as long as 7000 years ago but the beginning of cereal cultivation is less clear. Sub-Saharan Africa had a limited range of crops when Europeans first arrived in the 15th century, the most important being sorghum Sorghum vulgare and several millets. In parts of West Africa indigenous yams, rice and banana were grown. New food crops, including cassava and maize, were introduced after the discovery of the Americas.

The first cattle entered Africa through Egypt about 7000 years ago. The plough, by then present over most of the Old World, did not reach Africa until the 19th century except in Ethiopia.

Unlike Asia, animal and arable agriculture were generally separate in Africa and could have been a barrier to early intensification.

Even domesticated cattle could be local, because genetic evidence seems to suggest it is of a distinct strain than Eurasian cattle:

http://www.oxfordbibliographies.com/view/document/obo-9780199846733/obo-9780199846733-0115.xml


Early farming and pastoralism, or food production, in Africa can be separated into several categories: animals, grains, and tropical plants, all of which prevailed in different places and at different times.

Animal domestication is the earliest recorded, but is highly disputed. Large cattle bones found in Egypt dated to the 10th millennium BP are deemed to be domestic on the basis that they could not have survived without human intervention and were found associated with pottery. The alternative view is that the timing is such that these cattle were wild, and that domestic cattle that arrived in the 8th millennium BP were derived from different Levantine stock. The waters are further muddied by the genetics of African cattle suggesting an independent strain, but this also has its critics.

With the general drying of the Sahara around 5000 BP, herders and their cattle and small stock moved south with the tsetse belts into West and East Africa, and by 2000 BP had reached Southern Africa. The question of hunters becoming food producers without apprenticeship is debated, as is the concept of a Stone Age pastoral or agricultural “Neolithic” in Africa. Although winter rainfall crops, such as wheat and barley, were used in Dynastic Egypt, domestication of grain outside the Nile Valley was considerably later than that of animals, only occurring in the Sahel c. 3800 BP, although wild grains most probably had been collected by herders long before this. The beginnings of tropical plant domestication are more difficult to see, as preservation has made the plant residues hard to find. These plants include yams, rice, and oil-bearing trees. Often, environmental change, such as forest clearing, has to be used as a proxy for farming activities in tropical zones. In addition, the development of iron technology is closely correlated with the spread of farming societies in sub-Saharan Africa after 3000 BP. The history of food production in Africa lags somewhat behind the research done in the Near East and Europe, but genomic work on modern Africans has started in parallel with advanced linguistic work. Ancient DNA will be the next technological input now that the problems of contamination have been successfully addressed.

jose luis
31-01-20, 19:51
https://www.cell.com/cell/fulltext/S0092-8674(20)30059-3


New paper on this matters