PDA

View Full Version : Late European Neanderthal's mt-DNA might have been L2'3'4'5'6 derived



jeanL
11-13-2013, 04:45 PM
This is big!!! Like really big!!!!

Mitochondrial DNA Polymorphisms Shared between Modern Humans and Neanderthals: Adaptive Convergence or Evidence for Interspecific Hybridization? (http://download.springer.com/static/pdf/897/art%253A10.1134%252FS1022795413080115.pdf?auth66=1 384533610_5bd4342eefab3e5934b1cdf4a235e984&ext=.pdf)




Abstract—An analysis of the variability of the nucleotide sequences in the mitochondrial genome of modern humans, neanderthals, Denisovans, and other primates has shown that there are shared polymorphisms at positions 2758 and 7146 between modern Homo sapiens (in phylogenetic cluster L2'3'4'5'6) and Homo neanderthalensis (in the group of European neanderthals younger than 48 000 years).It is suggested that the convergence may be due to adaptive changes in the mitochondrial genomes of modern humans and neanderthals or interspecific hybridization associated with mtDNA recombination.

919

Jean M
11-13-2013, 05:04 PM
It would be very big if it really provided evidence of interbreeding. But the proposed basis of that would be a blow to those who wish to see Neanderthals as the same species as Homo sapiens.


during interspecific hybridization, the leakage of the paternal mtDNA is possible due to the fact that the system of eliminating the paternal mtDNA (along with the sperm cytoplasm) is strictly speciesspecific and may fail at interspecific hybridization. Although cases of interspecific hybridization are rare, they are described in the literature for different groups of animals, including mammals

They conclude that convergent adaptation is more likely:


Therefore, it is likely that the convergence of the A415T amino acid variant of cytochromeсoxidase 1 in the mtDNA phylogenetic groups of modern humans and Neanderthals may have arisen as a result of adaptation to the changing conditions of the environment. In summary, it should be noted that the question of whether the convergence of polymorphism variants in the coding region of mtDNA of modern humans and Neanderthals is a consequence of adaptive changes in their mitochondrial genomes or interspecific hybridization associated with mtDNA recombination remains open.

jeanL
11-13-2013, 05:11 PM
It would be very big if it really provided evidence of interbreeding. But the proposed basis of that would be a blow to those who wish to see Neanderthals as the same species as Homo sapiens.
They conclude that convergent adaptation is more likely:

Well the amino acid change of A415T would make sense, but we are talking about two distant SNPs in here, one in position 2758, one in 7146, so you can explain one as convergent evolution, but what about the other?? How about a third explanation that European Neanderthal's might have carried L2'3'4'5'6, and West Asian Neanderthal didn't carry them, and that L2'3'4'5'6 split from L1 and L0 a long time ago, and L2'3'4'5'6+ and that L2'3'4'5'6 are actually Neanderthal derived clades. How does this fit with the Out of Africa scenario?? Mind you, this isn't just one Neanderthal carry it, but all European Neanderthal carrying it, from Croatia to Germany to Spain.

Jean M
11-13-2013, 05:46 PM
How about a third explanation that European Neanderthal's might have carried L2'3'4'5'6, and West Asian Neanderthal didn't carry them, and that L2'3'4'5'6 split from L1 and L0 a long time ago, and L2'3'4'5'6+ and that L2'3'4'5'6 are actually Neanderthal derived clades.

Doesn't work at all. The Neanderthals do not carry L2'3'4'5'6. They simply have two substitutions in common with L2'3'4'5'6. We cannot have a scenario in which L2'3'4'5'6 in toto passed from Neanderthal to Homo sapiens. As the paper says, it has long been understood that Neanderthals and Homo sapiens are on different branchlines from a common ancestor. As the paper says, the only form of hybridization which could be postulated to create this effect of two substitutions in common is the (highly unlikely) leakage of the paternal mtDNA from some Neanderthal (who managed get far enough south to interbreed with L2'3'4'5'6).

German Dziebel
11-13-2013, 06:29 PM
Doesn't work at all. The Neanderthals do not carry L2'3'4'5'6. They simply have two substitutions in common with L2'3'4'5'6 .

No need to be so categorical. Under the hypothesis that L2'3'4'5'6 introgressed from Neandertals, the absence of all other mutations that define the L2'3'4'5'6 clade in modern humans on European Neandertal sequences only means that on modern humans they arose after the introgression. Whatever modern humans had before got replaced by this Neandertal-derived clade. And this clade migrated to Africa as the presence of a basal M1 lineage in Africa and a basal L6 lineage outside of Africa suggest. L1 and L0 are out of the picture and may have also introgressed into modern Afrcian populations from archaic Africans.

R.Rocca
11-13-2013, 09:17 PM
.No need to be so categorical. Under the hypothesis that L2'3'4'5'6 introgressed from Neandertals, the absence of all other mutations that define the L2'3'4'5'6 clade in modern humans on European Neandertal sequences only means that on modern humans they arose after the introgression. Whatever modern humans had before got replaced by this Neandertal-derived clade. And this clade migrated to Africa as the presence of a basal M1 lineage in Africa and a basal L6 lineage outside of Africa suggest. L1 and L0 are out of the picture and may have also introgressed into modern Afrcian populations from archaic Africans.

Can you tell us what your definition of genetic introgression is?

Also, are you saying that events at 2758 and 7146 for the base of one of your trees?

parasar
11-13-2013, 09:39 PM
Introgression, I believe, is the transfer of genetic material outside the normal mtDNA and nuclear DNA inheritance process whereby in a introgressive hybridization event a sequence is grafted onto another.

German Dziebel
11-13-2013, 10:46 PM
Can you tell us what your definition of genetic introgression is?

I keep the exact content of the term "introgression" open. It depends on how we end up treating Neandertals or other archaic populations - as extinct populations within our own species or as different species. In the former case, introgression would mean simply admixture between two previously isolated populations (recall Polynesians some of whom are almost fixed at Y-DNA C2 but it's an ancient lineage they picked up from a replaced population in Near Oceania and not an authentic Asian lineage); in the latter a more complex process of hybridization with backcrossing.


Also, are you saying that events at 2758 and 7146 for the base of one of your trees?

According to Malyarchuk, these two mutations are present on all sequences classified as L2'3'4'5'6. So, yes, we can put the sequences of European Neandertals carrying these 2 mutations on the left and all human L2'3'4'5'6 sequences on the right. Neandertals being a Eurasian species, this would place the root of this tree outside of Africa and will make most of M and N as well as L6a the remnants of that Eurasian population that stayed behind and didn't colonize Africa. Hgs M1, L2, L3, L4, L5, L6b will be the lineages that migrated to Africa.

But I'm not 100% sure that hg R is part of that introgressed clade... Technically it is, but if we put Y-DNA and mtDNA side by side, we'll see that mtDNA R distributionally corresponds to Y-DNA CF and its typological position is different from the position of R - it's not a subset of a subset of Eurasian lineages.

lgmayka
11-14-2013, 12:14 AM
Introgression, I believe, is the transfer of genetic material outside the normal mtDNA and nuclear DNA inheritance process whereby in a introgressive hybridization event a sequence is grafted onto another.
Introgression initially meant a specific kind of selective breeding across species (http://en.wikipedia.org/wiki/Introgression), particularly plants:
---
Introgression, also known as introgressive hybridization, in genetics (particularly plant genetics) is the movement of a gene (gene flow) from one species into the gene pool of another by the repeated backcrossing of an interspecific hybrid with one of its parent species.
---

If we extrapolate the definition to humans, and to natural selection rather than guided breeding, we end up with the notion of a specific highly advantageous mutation that jumps from one population group to another, in a 2-step process of hybridization followed by natural selection. An example is described in this paper (http://www.pnas.org/content/103/48/18178.long): "Evidence that the adaptive allele of the brain size gene microcephalin introgressed into Homo sapiens from an archaic Homo lineage".

Population geneticists extend the defintion even further to refer to a specific genetic pattern which has jumped from one population group to another by any means--i.e., with or without the operation of natural selection. So for example, a population geneticist might claim that R1a-CTS6 (the so-called Ashkenazi Levite clade) introgressed into the ancient or medieval Jewish community from another group.

But note that introgression does not imply any kind of "abnormal" inheritance or "grafting." Introgression is the statistical result of normal genetic inheritance, modified by either natural or sentient selection.

Jean M
11-14-2013, 12:33 AM
An example is described in this paper (http://www.pnas.org/content/103/48/18178.long): "Evidence that the adaptive allele of the brain size gene microcephalin introgressed into Homo sapiens from an archaic Homo lineage".


Certainly wasn't from Neanderthals. They didn't carry the variant D.

As I recall that paper was one several guessing what wonderful things we might have inherited from Neanderthals which turned out incorrect on the publication of the Neanderthal genomes.

parasar
11-14-2013, 12:57 AM
[URL="http://en.wikipedia.org/wiki/Introgression"]...
But note that introgression does not imply any kind of "abnormal" inheritance or "grafting." Introgression is the statistical result of normal genetic inheritance, modified by either natural or sentient selection.

I agree.

The Malyarchuk et. al. scenario would be specific case of an out of norm example.

parasar
11-14-2013, 01:16 AM
...
Neandertals being a Eurasian species, this would place the root of this tree outside of Africa ...
But I'm not 100% sure that hg R is part of that introgressed clade... Technically it is, but if we put Y-DNA and mtDNA side by side, we'll see that mtDNA R distributionally corresponds to Y-DNA CF and its typological position is different from the position of R - it's not a subset of a subset of Eurasian lineages.

Since Malyarchuk et. al. propose an introgression from a modern human male mtDNA to a European Neanderthal female, wouldn't only the introgression event have to be outside Africa?

Gisele H
11-14-2013, 03:21 AM
... if we put Y-DNA and mtDNA side by side, we'll see that mtDNA R distributionally corresponds to Y-DNA CF....

I am not following you here... Most Asian mtDNA haplogroups are not R.

GailT
11-14-2013, 05:21 AM
According to Malyarchuk, these two mutations are present on all sequences classified as L2'3'4'5'6. So, yes, we can put the sequences of European Neandertals carrying these 2 mutations on the left and all human L2'3'4'5'6 sequences on the right. Neandertals being a Eurasian species, this would place the root of this tree outside of Africa and will make most of M and N as well as L6a the remnants of that Eurasian population that stayed behind and didn't colonize Africa. Hgs M1, L2, L3, L4, L5, L6b will be the lineages that migrated to Africa.


There are several reasons why this doesn't work. First, you can't build a phylogenetic tree from two mutations in isolation. You also have to look at the hundreds of mutations that distinguish the European branch of Neandertals from L2'3'4'5'6, This is shown very clearly in Behar et al 2012, in Figure 1 on page 2 (link). (http://ac.els-cdn.com/S0002929712001462/1-s2.0-S0002929712001462-main.pdf?_tid=0d0d5724-4ce8-11e3-9437-00000aab0f26&acdnat=1384404708_011b471eb08692c1f42cb14b832a545d )

The most likely way these two specific mutations could be shared is that they have been selected or they occurred as random mutations independently in each clade. If they are silent mutations, random mutations is more likely.

Introgression is not very plausible, as you would have to imagine a biological process in which only those two base pairs transferred from paternal mtDNA, replacing the two base pairs in the maternal mtDNA, but none of the intervening mutations between 2758 and 7146 were transferred. In other words, a recombination of paternal and maternal mtDNA in which only two among several hundred mutations were transferred from the paternal to maternal mtDNA.

Even if this had occurred, it would be evidence of admixture of Neandertals and modern humans. But the Neandertal maternal line became extinct, so there is still no evidence of Neandertal migration from Europe to Africa. But more importantly, the Neandertal mtDNA is very clearly not ancestral to L2'3'4'5'6, as shown in Figure 1 in Behar. Finally, if there were a scenario in which L2'3'4'5'6 originated in Eurasia, this still leaves the deepest mtDNA ancestry (L0 and L1) in Africa, and you have to explain why all other L haplogroups are found primarily in Africa today, not Eurasia.

German Dziebel
11-14-2013, 02:05 PM
I am not following you here... Most Asian mtDNA haplogroups are not R.

Regional frequencies can fluctuate (e.g., hg B in America was likely more frequent in North America in the past than now), but both Y-DNA CF and mtDNA R have the widest geographic distribution worldwide and are well attested in West Eurasia, Australia and South America.

German Dziebel
11-14-2013, 02:25 PM
First, you can't build a phylogenetic tree from two mutations in isolation..

What do you mean in "isolation"? And why cannot you build a tree with 2 root mutations? I gave you examples of 1-2 mutation-based clades on human mtDNA. Currently, L2'3'4'5'6 is separated from L1 by 5 mutations, while L0 has 10 defining mutations to it. By your logic, we shouldn't have the L1 node because it's only 5 mutations and not 10.


You also have to look at the hundreds of mutations that distinguish the European branch of Neandertals from L2'3'4'5'6, This is shown very clearly in Behar et al 2012, in Figure 1 on page 2 (link). (http://ac.els-cdn.com/S0002929712001462/1-s2.0-S0002929712001462-main.pdf?_tid=0d0d5724-4ce8-11e3-9437-00000aab0f26&acdnat=1384404708_011b471eb08692c1f42cb14b832a545d )

Behar didn't notice the sharing between European Neandertals and human lineages outside of L0 and L1. So at the moment I would hold off on using him as a gold standard?


The most likely way these two specific mutations could be shared is that they have been selected or they occurred as random mutations independently in each clade. If they are silent mutations, random mutations is more likely.

You are using the word "likely" idly. I don't know what's more likely. It's of course possible that what Malyarchuk found is convergence, and he admits this, but let's not jump the gun and try to kill what may be evidence of something more important. An admixture scenario makes sense in light of the whole-genome data. All non-Africans and some Sub-Saharan Africans were showed to carry Neandertal-derived alleles. This is the bigger picture. Plus archaic admixture would explain all the phylogeographic problems with the current tree.


Even if this had occurred, it would be evidence of admixture of Neandertals and modern humans. But the Neandertal maternal line became extinct, so there is still no evidence of Neandertal migration from Europe to Africa. But more importantly, the Neandertal mtDNA is very clearly not ancestral to L2'3'4'5'6, as shown in Figure 1 in Behar. Finally, if there were a scenario in which L2'3'4'5'6 originated in Eurasia, this still leaves the deepest mtDNA ancestry (L0 and L1) in Africa, and you have to explain why all other L haplogroups are found primarily in Africa today, not Eurasia.

I've already explained it multiple times: L0 and L1 introgressed into modern Africans (who had previously absorbed Neandertal DNA in Eurasia) from archaic Africans. That's why L0 and L1 are so divergent, low-frequency and African-specific. While Neandertals didn't migrate to Africa, modern humans who carried their genes did.

German Dziebel
11-14-2013, 02:32 PM
Since Malyarchuk et. al. propose an introgression from a modern human male mtDNA to a European Neanderthal female, wouldn't only the introgression event have to be outside Africa?

As Gisele pointed out in an e-mail to me, it doesn't look like Malyarchuk has made up his mind. He writes, "Thus, if one adheres to the hybrid theory, the appearance of this combination of SNP variants in European Neanderthals could be due to the hybridization between female individuals of Neanderthals and male sapiens, which occurred after the appearance of CroMagnons in Europe (40000–50000 years ago). The results of recent genome research showed that one of the last episodes of interspecific hybridization could occur in
Europe, most probably, in the range from 47000–65000 years ago."

We can always imagine a scenario whereby L0 and L1 people migrated out of Africa to Europe, admixed with Neandertal women there, got their L0 and L1 haplotypes replaced by a new clade dominant and then migrated back to Africa as L2'3'4'5'6. It's a possibility, true.

Jean M
11-14-2013, 02:59 PM
We can always imagine a scenario whereby L0 and L1 people migrated out of Africa to Europe, admixed with Neandertal women there, got their L0 and L1 haplotypes replaced by a new clade dominant and then migrated back to Africa as L2'3'4'5'6.

You might be able to imagine this scenario. Those of us burdened by evidence and tiresomely wedded to logical deduction have a little more difficulty. There is no trail of AMH into Europe and back to Africa at the time proposed for L0. On the evidence before us, AMH first entered Europe c. 46,000 years ago, and were carrying mtDNA U.

German Dziebel
11-14-2013, 03:23 PM
You might be able to imagine this scenario. Those of us burdened by evidence and tiresomely wedded to logical deduction have a little more difficulty. There is no trail of AMH into Europe and back to Africa at the time proposed for L0. On the evidence before us, AMH first entered Europe c. 46,000 years ago, and were carrying mtDNA U.

I was just trying to give out-of-Africa a chance. In my opinion, it has no chances in light of the growing ancient DNA data and all other cross-interdisciplinary evidence that we have. I still have to see evidence of your mastery of facts and logic related to modern human origins.

AJL
11-14-2013, 04:55 PM
I was just trying to give out-of-Africa a chance. In my opinion, it has no chances in light of the growing ancient DNA data and all other cross-interdisciplinary evidence that we have. I still have to see evidence of your mastery of facts and logic related to modern human origins.

????

Do you even know what DNA and fossil data there is and what it suggests? Or will you simply restate what to my eyes seems only to be overweening pride in the extreme minority position you hold for an Out-of-Americas origin, regardless of all available DNA and fossil evidence?

zaender
11-14-2013, 05:43 PM
from the very beginning of comparing there had been the average difference betbeen modern humans of 8 mutation on HVR1 level and that of neanderthals to humans was 27 mutations in average. So, like Gail already said, on a CR-level we encounter hundreds of differences and noone will reach to push any neanderthal under the node of L In contrary, we probably will find more AMHs with ancient mtdna like Lake Mungo3, who precede the origin of L.

vettor
11-14-2013, 06:29 PM
@german dziebel

IIRC my studies, it is stated that , humans can impregnate a neanderthal, but a neanderthal cannot impregnate a human. Where does your theory stand in regards to this?

parasar
11-14-2013, 07:43 PM
@german dziebel

IIRC my studies, it is stated that , humans can impregnate a neanderthal, but a neanderthal cannot impregnate a human. Where does your theory stand in regards to this?

I don't understand this stuff fully, but as far I could understand, both ways works. In fact for Densiovans who (apparently - as there is some controversy) diverged even prior to Neanderthals, the introgression is theorized to be from Denisovan males to modern human females, based on the lower Denisovan seen on the X.

GailT
11-14-2013, 09:10 PM
What do you mean in "isolation"? And why cannot you build a tree with 2 root mutations? I gave you examples of 1-2 mutation-based clades on human mtDNA. Currently, L2'3'4'5'6 is separated from L1 by 5 mutations, while L0 has 10 defining mutations to it. By your logic, we shouldn't have the L1 node because it's only 5 mutations and not 10.

I mean that you cannot construct a tree using 2 mutations that fit your model while ignoring hundreds of mutations that contradict your model. You must consider the entire set of mutations found in each sample, and if you to do this, you'll see that your version of the tree is very obviously contradicted by the data.



Behar didn't notice the sharing between European Neandertals and human lineages outside of L0 and L1. So at the moment I would hold off on using him as a gold standard?

Behar et al. did notice the sharing of these mutations. If you look at Figure 1 in the Behar et al. paper, you will notice that both of these mutations are underlined, as are several other mutations that are found in multiple branches of there tree. So they developed the most parsimonius tree that accounted for the fact that some mutations occurred independently in multiple branches.




You are using the word "likely" idly. I don't know what's more likely. It's of course possible that what Malyarchuk found is convergence, and he admits this, but let's not jump the gun and try to kill what may be evidence of something more important. An admixture scenario makes sense in light of the whole-genome data. All non-Africans and some Sub-Saharan Africans were showed to carry Neandertal-derived alleles. This is the bigger picture. Plus archaic admixture would explain all the phylogeographic problems with the current tree.

I'm not using the word "likely" idly. There are many cases in which the same mutation occurs independently in different branches of the tree, so this explanation is reasonable and possible, while invoking a new mechanism of hydbridization of mtDNA sounds very much like special pleading. In any case, regardless of the source of these two mutations, it does not affect the mtDNA tree, as is readily apparent in Figure 1 of Behar et al. So even if hybridization did occur and two mutations were transferred, the most parsimonius tree does not change because it is based on an analysis of hundreds of mutations, not just two.



I've already explained it multiple times: L0 and L1 introgressed into modern Africans (who had previously absorbed Neandertal DNA in Eurasia) from archaic Africans. That's why L0 and L1 are so divergent, low-frequency and African-specific. While Neandertals didn't migrate to Africa, modern humans who carried their genes did.

That "explanation" is obviously flawed. L0 and L1 did not introgress into modern humans; they are modern humans L0 and L1 have age estimates of around 130,000 years, long after the evolution of modern humans. Modern humans do not carry Neandertal mtDNA, and there is very little evidence of Neandertal autosomal DNA in modern Africans. Autosomal DNA from Neandertals is found at very low percentages in people outside of Africa, indicating admixture after modern humans left Africa. Eurasian specific mtDNA haplogroups are less than 70,000 years old. The specific mtDNA haplogroups that did migrate back to Africa are much younger than date, around 30,000 years, You keep making these claims, but where is your evidence? What phylotrees have you constructed that are consistent with the mtDNA data and that support your theory? I think if you actually try to build a mtDNA Phylotree, you will see that it is inconsistent with your theory.

German Dziebel
11-27-2013, 01:44 AM
I mean that you cannot construct a tree using 2 mutations that fit your model while ignoring hundreds of mutations that contradict your model. You must consider the entire set of mutations found in each sample, and if you to do this, you'll see that your version of the tree is very obviously contradicted by the data.

There are dozens of sites at which individual human mtDNA sequences match Denisovans and/or Neandertals to the exclusion of other human sequences. How is this data reflected in the current tree?


Behar et al. did notice the sharing of these mutations. If you look at Figure 1 in the Behar et al. paper, you will notice that both of these mutations are underlined, as are several other mutations that are found in multiple branches of there tree. So they developed the most parsimonius tree that accounted for the fact that some mutations occurred independently in multiple branches.

Yes, he did. My bad. But his "most parsimonious tree" ignores available ancient DNA data. He retrofitted the human mtDNA tree constructed prior to the appearance of ancient DNA evidence instead of building it using ancestral states now known from Neandertal and/or Denisovan mtDNAs. Of course, some mutations occur independently in multiple branches but not those that are clearly identical with Denisovan/or Nenadertal alleles. Those are related to the human ones by descent. They are not homoplastic.


I'm not using the word "likely" idly. There are many cases in which the same mutation occurs independently in different branches of the tree, so this explanation is reasonable and possible, while invoking a new mechanism of hydbridization of mtDNA sounds very much like special pleading. In any case, regardless of the source of these two mutations, it does not affect the mtDNA tree, as is readily apparent in Figure 1 of Behar et al. So even if hybridization did occur and two mutations were transferred, the most parsimonius tree does not change because it is based on an analysis of hundreds of mutations, not just two.

See above. There are many more cases like the one highlighted by Malyarchuk.


L0 and L1 did not introgress into modern humans; they are modern humans L0 and L1 have age estimates of around 130,000 years, long after the evolution of modern humans.

We don't know that. All those "anatomically modern human" remains in Africa haven't yielded any DNA, and all the skulls of presumably behaviorally modern humans in Africa (Hofmeyr), Asia (Zhoukoudian), America (Lagoa Santa) or Europe cluster together and the African ones are not closer to the AMH skulls in Africa.


Modern humans do not carry Neandertal mtDNA, and there is very little evidence of Neandertal autosomal DNA in modern Africans.

If it's "little" by autosomal standards, it can be significant at a certain single locus such as mtDNA.

GailT
11-27-2013, 01:58 AM
Yes, he did. My bad. But his "most parsimonious tree" ignores available ancient DNA data. He retrofitted the human mtDNA tree constructed prior to the appearance of ancient DNA evidence instead of building it using ancestral states now known from Neandertal and/or Denisovan mtDNAs.

Behar et al 2012 used several Neandertal mtDNA samples. They are included in the most parsimonious tree. There are no data that show mtDNA descent of modern humans from Neandertals or Denisovans. If you actually had such data, you would overturn the scientific consensus of the last 25 years and you could publish your findings in your journal of choice, Science, Nature, where ever you like. Claiming on a hobbyist discussion forum that the entire scientific community is wrong, without providing any data to support the claim, is not very productive.

German Dziebel
11-27-2013, 03:13 AM
Behar et al 2012 used several Neandertal mtDNA samples. They are included in the most parsimonious tree. There are no data that show mtDNA descent of modern humans from Neandertals or Denisovans. If you actually had such data, you would overturn the scientific consensus of the last 25 years and you could publish your findings in your journal of choice, Science, Nature, where ever you like. Claiming on a hobbyist discussion forum that the entire scientific community is wrong, without providing any data to support the claim, is not very productive.

Agree. We got too far here. But the hobbyist community should know that the science is not a bulletproof monolith handed down from an enlightened minority. And out-of-Africa (with all its assets such as mtDNA phylogeny or AMH remains in Africa) is a model that needs to be tested and debated, not a fact to be assumed.