Tweet
[Note: I used samples from ysearch.org to investigate what I wrote below, so this may be particularly interesting for some ftdna members.]
In 2007 a genetic survey of England found a very weird haplogroup in a man from Yorkshire. It was A1a (back then it was known as A1), which is found mainly in West Africa. The geneticists compared the haplotype of the Yorkshire A1a with A1a samples from Guinea-Bissau, Niger, and an African American and concluded that, though it wasn't very typical of sub-Saharan A1a, it was close enough to connect it to a historic origin from Africa. The study can be found here (it's freely available):
When the geneticists studied A1a they had access to 11 samples. But the world of human genetics is exploding. I found 23 haplotypes, and with this new information I have grouped the A1a samples into 3 clusters:
[See attached image at end of post: A1a-haplotypetable.gif]
The following graph neatly illustrates the cluster membership by showing the genetic distances between the clusters. Note that the graph claims to show the distance from the modals of each cluster but it actually shows the distance from the average values of each cluster.
[See attached image at end of post: A1a-geneticdistancesbetweenclusters.gif]
I was very thorough about eliminating related samples and making sure that any samples I predicted as belonging to A1a really did belong to A1a. I seriously doubt I made a mistake, but of course, it's not impossible. There are only 2 samples that have an exact 12-marker match, but they're from Belgium and England, plus one was self-reported via Family Tree DNA and the other was submitted to yhrd.org, so it probably came from a genetic survey. Also, 12-marker matches aren't uncommon amongst unrelated people belonging to the same haplogroup. 2 samples came from the same city, Leuven, Belgium, but their genetic distance is one of the largest of any 2 samples in the Yorkshire cluster, and since they originate from yhrd.org, they probably came from a scientific survey, and scientific surveys try to ensure none of their testees are related.
Regarding their geographic origin, the Yorkshire cluster is found in England, Belgium, and Niger (I think it's possible the Niger samples actually came from Mali, but it's not too important), the Sahel cluster is found in Guinea-Bissau, Senegal and Niger (another Niger sample from the same population as the previous one), and the West Africa cluster doesn't have any samples that can be located indisputably because they were all reported in people born in the US. But on account of the fact that 2 of these samples came from African Americans and that the other 4 are more likely than not to come from African Americans, I am presuming this cluster corresponds to the rest of West Africa.
[See attached image at end of post: A1a-geographicrangeEurope.gif]
[See attached image at end of post: A1a-geographicrangeAfrica.gif]
There would seem to be 2 samples that don't fit. There's a European American belonging to the Sahel cluster and a Nigeri (not Nigerian) belonging to the Yorkshire cluster. I have a theory how this could have come about. The highest percentage of A1a is in the Guinean region. All these samples belong to the Sahel cluster and they have by far the highest diversity of the 3 clusters. I think this is where A1a originated, with the Sahel modal. It diffused eastwards into the rest of West Africa, where it transmuted into the West Africa modal. It also diffused northwards to North Africa and Europe. The presence of an A1a of the Sahel cluster in a European American would be due to this ancient diffusion. The A1a in North Africa then transmuted into the Yorkshire modal and then spread into Europe and the Sahara. This would explain the weird distribution of the Yorkshire cluster, found in Northwest Europe and the southern part of the Sahara. To settle this mistery, we need to haplotype the known A1a samples from Morocco. There's also 3 A samples from Portugal and the Azores, and an apparent A sample from Germany. But a supposed A(xA3b2) from South Tyrol (Italy) almost certainly belongs to J and was misreported.
Due to the lack of representation of the central region of the range of A1a in the 2007 study, their discovery of a close match between the Yorkshire A1a and the Nigeri A1a was inevitably interpreted as representing a historic diffusion of A1a from Africa to Europe. There was already a slight hint that things were amiss when they stated that: "Although the British haplotype is peripheral, it lies equidistant (four mutational steps) from Niger and Guinea-Bissau haplotypes". Given the limited samples they had to work with, they reached the correct conclusion. The true mistake was not including A1a samples from a middle region like North Africa. But they knew about the A1a in Morocco, so if they didn't test them it's probably because the samples had already been destroyed.
I also have a comment about the very curious distribution of the Yorkshire A1a, found on either side of the English Channel, but it's very controversial. As I've noted several times before, I don't believe one comma about the whole TMRCA theory, and I think almost all haplogroups have existed for tens of thousands of years. The Yorkshire A1a would have been scattered across the dry land between England and Holland, and the end of the Ice Age split its distribution into 2 camps on either side of the English Channel. In fact, the reason I started looking into A1a is because of my post a few weeks ago posing a challenge to believers of the TMRCA theory. In order to counter the responses I got to that post, I needed to find STR data on some very exotic haplogroups, such as A1, D2, or E1a. As I should have expected, I became fascinated in trying to decypher the haplogroups' composition and got completely sidetracked. And I'm still not done, because I also found some very interesting things about A3b2, E1a, and E2, and I'm going to write about them and post that later on.
In 2007 a genetic survey of England found a very weird haplogroup in a man from Yorkshire. It was A1a (back then it was known as A1), which is found mainly in West Africa. The geneticists compared the haplotype of the Yorkshire A1a with A1a samples from Guinea-Bissau, Niger, and an African American and concluded that, though it wasn't very typical of sub-Saharan A1a, it was close enough to connect it to a historic origin from Africa. The study can be found here (it's freely available):
When the geneticists studied A1a they had access to 11 samples. But the world of human genetics is exploding. I found 23 haplotypes, and with this new information I have grouped the A1a samples into 3 clusters:
[See attached image at end of post: A1a-haplotypetable.gif]
The following graph neatly illustrates the cluster membership by showing the genetic distances between the clusters. Note that the graph claims to show the distance from the modals of each cluster but it actually shows the distance from the average values of each cluster.
[See attached image at end of post: A1a-geneticdistancesbetweenclusters.gif]
I was very thorough about eliminating related samples and making sure that any samples I predicted as belonging to A1a really did belong to A1a. I seriously doubt I made a mistake, but of course, it's not impossible. There are only 2 samples that have an exact 12-marker match, but they're from Belgium and England, plus one was self-reported via Family Tree DNA and the other was submitted to yhrd.org, so it probably came from a genetic survey. Also, 12-marker matches aren't uncommon amongst unrelated people belonging to the same haplogroup. 2 samples came from the same city, Leuven, Belgium, but their genetic distance is one of the largest of any 2 samples in the Yorkshire cluster, and since they originate from yhrd.org, they probably came from a scientific survey, and scientific surveys try to ensure none of their testees are related.
Regarding their geographic origin, the Yorkshire cluster is found in England, Belgium, and Niger (I think it's possible the Niger samples actually came from Mali, but it's not too important), the Sahel cluster is found in Guinea-Bissau, Senegal and Niger (another Niger sample from the same population as the previous one), and the West Africa cluster doesn't have any samples that can be located indisputably because they were all reported in people born in the US. But on account of the fact that 2 of these samples came from African Americans and that the other 4 are more likely than not to come from African Americans, I am presuming this cluster corresponds to the rest of West Africa.
[See attached image at end of post: A1a-geographicrangeEurope.gif]
[See attached image at end of post: A1a-geographicrangeAfrica.gif]
There would seem to be 2 samples that don't fit. There's a European American belonging to the Sahel cluster and a Nigeri (not Nigerian) belonging to the Yorkshire cluster. I have a theory how this could have come about. The highest percentage of A1a is in the Guinean region. All these samples belong to the Sahel cluster and they have by far the highest diversity of the 3 clusters. I think this is where A1a originated, with the Sahel modal. It diffused eastwards into the rest of West Africa, where it transmuted into the West Africa modal. It also diffused northwards to North Africa and Europe. The presence of an A1a of the Sahel cluster in a European American would be due to this ancient diffusion. The A1a in North Africa then transmuted into the Yorkshire modal and then spread into Europe and the Sahara. This would explain the weird distribution of the Yorkshire cluster, found in Northwest Europe and the southern part of the Sahara. To settle this mistery, we need to haplotype the known A1a samples from Morocco. There's also 3 A samples from Portugal and the Azores, and an apparent A sample from Germany. But a supposed A(xA3b2) from South Tyrol (Italy) almost certainly belongs to J and was misreported.
Due to the lack of representation of the central region of the range of A1a in the 2007 study, their discovery of a close match between the Yorkshire A1a and the Nigeri A1a was inevitably interpreted as representing a historic diffusion of A1a from Africa to Europe. There was already a slight hint that things were amiss when they stated that: "Although the British haplotype is peripheral, it lies equidistant (four mutational steps) from Niger and Guinea-Bissau haplotypes". Given the limited samples they had to work with, they reached the correct conclusion. The true mistake was not including A1a samples from a middle region like North Africa. But they knew about the A1a in Morocco, so if they didn't test them it's probably because the samples had already been destroyed.
I also have a comment about the very curious distribution of the Yorkshire A1a, found on either side of the English Channel, but it's very controversial. As I've noted several times before, I don't believe one comma about the whole TMRCA theory, and I think almost all haplogroups have existed for tens of thousands of years. The Yorkshire A1a would have been scattered across the dry land between England and Holland, and the end of the Ice Age split its distribution into 2 camps on either side of the English Channel. In fact, the reason I started looking into A1a is because of my post a few weeks ago posing a challenge to believers of the TMRCA theory. In order to counter the responses I got to that post, I needed to find STR data on some very exotic haplogroups, such as A1, D2, or E1a. As I should have expected, I became fascinated in trying to decypher the haplogroups' composition and got completely sidetracked. And I'm still not done, because I also found some very interesting things about A3b2, E1a, and E2, and I'm going to write about them and post that later on.
.
Comment