PDA

View Full Version : hg T in Somalis. Could it have come with Islam.



Almagest
08-09-2017, 09:30 PM
Sanchez (2005) suggests a E-V32 TMRCA of 4-4.8 kya in Somalis (n=151). With SNP testing (n=4) we know this number is closer to 3.5-4 kya. He also suggests a TMRCA of 2.1-2.5 kya for hg T Somalis (n=21). This could also be an over estimation.

Here is the TMRCA for hg T Somalis from Tillmar (2009) (n=12), using the infinite alleles model, average STR mutation rate of 0.022. The lower bound is an estimate for a 25 year generation and the upper bound is using 30 year generation.

Tillmar (2009) (n=12)

TMRCA = 2.6-3.1kya. This includes a haplotype that is far removed from all the others so the estimate is skewed. As an example, his DYS385ab 16-16 result is not found in Y*(xE3b1) group in the Sanchez study, or the FTDNA database. It's not common in non Somali hg T members, or in general. When he is removed,

TMRCA = 1.5-2.3kya.



And the users in the FTDNA database (n=4) using the same method,

TMRCA = 1.2-1.8kya.

I have also seen some hg T members here believe that T in the Horn of Africa is not very old.

Awale
08-10-2017, 09:32 AM
Sanchez (2005) suggests a E-V32 TMRCA of 4-4.8 kya in Somalis (n=151). With SNP testing (n=4) we know this number is closer to 3.5-4 kya. He also suggests a TMRCA of 2.1-2.5 kya for hg T Somalis (n=21). This could also be an over estimation.

Yeah, those TMRCAs are interesting but I'll be intrigued to see further analyses in the future. I was also surprised by TMRCA this E-V32 subclade (http://www.anthrogenica.com/showthread.php?11485-Somalis-and-E-Y17859) has with an Egyptian at Yfull, honestly. At any rate, there's really very little chance this is owed to "Islamic period admixture" given how non-existent such admixture mostly looks in terms of autosomal DNA and mtDNA across Greater Somalia and how T appears even in hinterland areas of the Horn like Jijiga or Dire Dawa at high frequencies. I've also touched upon T and how it's found all over the Horn in the past as well:

http://www.anthrogenica.com/showthread.php?9417-Cushitic-haplogroup&p=235697&viewfull=1#post235697

Y-DNA T-M70 clearly looks like a pretty old lineage in our region. It's highly doubtful that it's found even in South-Erythraeic speakers but Somalis in particular coincidentally just got it recently from the outside and show no proof of this in terms of auDNA and mtDNA.


TMRCA = 2.6-3.1kya.
TMRCA = 1.5-2.3kya.
TMRCA of 2.1-2.5 kya
TMRCA = 1.2-1.8kya.

All but one of these datings (the last one) predate the Islamic period as well. And pretty much of all of them long predate when some of the Somali clans claim Arabian migrants founded their clans. The last dating also only comes close to when the Islamic period was just about beginning rather than being smack dab in the middle of it or something which would be more ideal.

vettor
08-14-2017, 05:52 PM
The T-m70 in east africa came mostly from the DIR Clan off Ethiopia who was 100% T-M70 ( unsure if it is T1a1 or T1a3 branch)

http://www.sciencedirect.com/science/article/pii/S1872497316302484

the highest ST values were observed for pair-wise comparisons involving the
ethnic groups characterized by the lowest intra-population diversity parameters, i.e. the
Saho from Eritrea and the Somali from Djibouti (pairwise ST values range 0.547 0.713
and 0.074 0.713, respectively, Supplementary Table 8). Both these groups are
characterized by high frequencies of binary haplogroups which are relatively uncommon in
other African populations (Supplementary Fig. 1, [13, 15, 17, 18, 46]), such as haplogroup
E-V22 in the Saho (88%) and T-M70 in the Somali from Djibouti (74% in the population,
100% in the Dir clan), with Y-STRs presenting a low degree of within-population molecular
differentiation (Supplementary Fig. 2B and 3). Overall, the observed high level of interpopulation
diversity highlights the fact it is important to take into account ethnic
composition and, possibly, the clan structure of the populations under study when making
forensic and evolutionary inferences.

as you can see by the Yfull tree for T-M184 ........the east-african T is not very old
https://en.wikipedia.org/wiki/File:Haplogroup_T-M184_tree.png

Almagest
08-14-2017, 08:58 PM
The T-m70 in east africa came mostly from the DIR Clan off Ethiopia who was 100% T-M70 ( unsure if it is T1a1 or T1a3 branch)

http://www.sciencedirect.com/science/article/pii/S1872497316302484

the highest ST values were observed for pair-wise comparisons involving the
ethnic groups characterized by the lowest intra-population diversity parameters, i.e. the
Saho from Eritrea and the Somali from Djibouti (pairwise ST values range 0.547 – 0.713
and 0.074 – 0.713, respectively, Supplementary Table 8). Both these groups are
characterized by high frequencies of binary haplogroups which are relatively uncommon in
other African populations (Supplementary Fig. 1, [13, 15, 17, 18, 46]), such as haplogroup
E-V22 in the Saho (88%) and T-M70 in the Somali from Djibouti (74% in the population,
100% in the Dir clan), with Y-STRs presenting a low degree of within-population molecular
differentiation (Supplementary Fig. 2B and 3). Overall, the observed high level of interpopulation
diversity highlights the fact it is important to take into account ethnic
composition and, possibly, the clan structure of the populations under study when making
forensic and evolutionary inferences.

as you can see by the Yfull tree for T-M184 ........the east-african T is not very old
https://en.wikipedia.org/wiki/File:Haplogroup_T-M184_tree.png

One guy is T-Y16897, negative for everything downstream. It is likely all Somalis given the short TMRCA belong to this sub clade.

vettor
08-14-2017, 11:10 PM
One guy is T-Y16897, negative for everything downstream. It is likely all Somalis given the short TMRCA belong to this sub clade.

Due to the high percentage of the DIR clan for T1a-m70, this can only mean a founder marker

drobbah
08-15-2017, 02:45 AM
Due to the high percentage of the DIR clan for T1a-m70, this can only mean a founder marker

The Garxajis subclan of the Isaaq confederation are also T1a.We need more Somalis from all clans to test their Y-DNA but I have no doubt there might be smaller clans all over the Somali Peninsula that will come out T-M70.Founder effects are quite common among Somalis which is why majority are V32 (basically one massive ass founder effect lol).Some clans like the Sheikhaal from what i seen are generally E1b1b1c and this is again because of founder effect.

This should not be surprising since Somalis are a highly patriarchal nomadic society.

radioavdelning
01-15-2018, 04:12 PM
I share with a lot of Somalis on 23andme and I've not seen any Somalis around that aren't E-V32.

This makes me wonder if those geneticists keep tribes in mind when they conduct these studies. There's a huge difference in asking N Somalis from one clan and N Somalis from a variety of different clans.