Here's my first admixture run using Reference 3 for Harappa participants with FTDNA data.
You can see the participant results in a spreadsheet as well as their ethnic breakdowns and the reference population results.
Here's our bar chart and table. Remember you can click on the legend or the table headers to sort.
If the above interactive charts are not working, here's a static bar graph.
Since this is my first analysis of FTDNA data, I asked HRP0006 to provide me with his FTDNA results (HRP0093) too so they can be compared. Let's see how that turned out.
| HRP0006 | HRP0093 | ||
| C1 | S Asian | 49.31% | 49.06% |
| C2 | Onge | 14.13% | 13.70% |
| C3 | E Asian | 1.12% | 0.00% |
| C4 | SW Asian | 14.65% | 12.52% |
| C5 | European | 18.88% | 22.44% |
| C6 | Siberian | 0.00% | 0.78% |
| C7 | W African | 0.00% | 0.00% |
| C8 | Papuan | 0.54% | 0.01% |
| C9 | American | 1.35% | 1.48% |
| C10 | San/Pygmy | 0.00% | 0.00% |
| C11 | E African | 0.00% | 0.00% |
There are differences of up to 3% but generally the results are reasonably close.
HRP0095 and HRP0100 thought they had possible South Asian ancestry. That seems fairly unlikely at least in the last few generations since their Onge component is zero or very low.
Thank you for the update, Zack. The "S Asian," "SW Asian," and "European" component values respectively, for Harappa participants and reference populations with an apparent Iraqi nexus, along with presumed language(current or relatively recent tradition), sorted in descending order according to the "SW Asian" component:
Sample/POP S Asian SW Asian Euro Language
S. Iraq Arab 19% 64% 4% Arabic
Ref3 Iq Jews 23% 62% 13% Hebrew/Aramaic
Iq/Egy Jewish 21% 61% 15% Hebrew/Aramaic?
Assyrian Avg 25% 58% 16% Aramaic
S. Iq Mandaean 23% 58% 16% Arabic/Aramaic
C. Iraq Arab 25% 51% 17% Arabic
Iraqi Kurd 30% 48% 18% Kurdish
Paul, those percentages are misleading if you don't take into account the small African admixture especially seen in Arabs (by Arab, I always mean only Muslim Arabic speakers). I am not even mentioning the small Mongoloid admixture seen in various groups - especially Muslim - of West Asia.
S. Iq Mandaean 23% 58% 16% Arabic/Aramaic
Mandaeans only spoke Aramaic before the Arab conquests AFAIK, and not Arabic.
Onur, the Aramaic dialect of the Mandaeans (Mandaic)is similar to my Assyrian-Aramaic dialect. With us, it has remained a vernacular. For the Mandaeans, however, knowledge of Aramaic, outside of the Mandaean clergy, is uncommon.
I do not understand why you say the data I posted is misleading. It suggests a possible shared relationship between Iraqi Jews, Mandaeans, and Assyrians. The Iraqi Arabs are distinguished. As is the lone Iraqi Kurd. And, as you correctly point out, the differences can be further reinforced by referring to the various Asian and African components present (or absent) in the aforementioned populations. One can also distinguish, I should add, between populations displaying greater consistency across the predominant components, by referring to the various Asian,African, etc. minority components. For instance, the "E African" found in the one Mandaean, but not observed in the Assyrian K=11 average values (rounded to nearest whole %):
POP Sasia Onge Easia Swasi Euro Siber Wafri Papua Ameri SanPy Eafri
ASY 25% 0% 0% 58% 16% 0% 0% 0% 0% 0% 0%
MAN 23% 1% 0% 58% 16% 0% 0% 0% 1% 0% 2%
Of course, the Mandaean sample of one does not allow us to draw too broad a conclusion (if any) from the data.
Paul, if the Black African slave theory is correct, the Negroid components in Arabs should be from post-Islamic Black African slaves. So by removing their Negroid components and almost certainly post-Islamic Mongoloid components (probably from Turkic slaves and invaders) and taking the sum of their Caucasoid components as 99 or 100% (you can do the same for the other West Asian Muslim populations), you can make a better comparison.
you can do the same for the other West Asian Muslim populations
and also for the Negroid admixed Mandaean individual
Biharis HRP0003 and HRP0098 are quite close - the latter with a slightly higher SW. Asian.
Cf. HRP0032 who is also from Bihar.
A bit surprising to me, as well. I am not aware of any Hindu ancestor....off course there would have been some, but nobody told me 🙂 My parents migrated from Bihar to Pak, when they were kids (at the time of partition)
Thanks Zack for running this.
A bit surprising to me, as well. I am not aware of any Hindu ancestor....off course there would have been some
Are you joking? Your entire people were Hindus with Buddhist and Jain minorities (who were both converts from Hinduism anyway) before the spread of Islam in Bihar. So you are most probably fully descended from Hindus ultimately.
I have no illusions on this one. I was just talking about anecdotes.
BTW, why am i more similar to Bihari Birhaman than HRP0032 (Bihari Kayastha)?
That is not very surprising. Most Sunni converts were brahmans or rajputs. They are referred to as Afghan or Pathan.
The Shia were sometimes of Arab or Persian extraction. The Mughal, a very small group, were Persianized Turko-Mongols.
oral history generally doesn't go very far up most lineages. great-great grandparents is what i hear. i know i'm descended from hindus because i have great-grand parents who were born hindus. additionally, one of my grandparents had a 'surname' which denoted their hindu caste (sarkar = kayastha), and the family maintained a plausible profession for that caste (who knows the arrow of causality).
if you hindu ancestry is mostly 200-300 years in the past, you wouldn't remember it through oral history. *especially* if you didn't want to (muslims don't for obvious reasons, unless those hindus were very high caste).
That is not very surprising. Most Sunni converts were brahmans or rajputs. They are referred to as Afghan or Pathan.
no they weren't. that's even evident in the HAP samples of muslims.
Onur , Jains are not hindu converts. There are also folk religions like punjabi jathera(ancestor worship) and not forgetting animism.
Simranjits, even though Jainism is old enough to have obscure origins, its spread in South Asia largely happened during the recorded history of South Asia and certainly its spread was largely into the Hindu territories (as Hinduism was the dominant religion in most of South Asia from very early times). So your statement "Jains are not Hindu converts" is refuted by history.
Do you know where the HAP muslim samples are from?
UP "Shias and Sunnis are closer to upper caste ..."
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1855350/
Bihar: "The high caste Brahmins and Muslims of Bihar show close genetic affinity." http://books.google.com/books?id=l2WZ4JF4zoMC&pg=PA52
"The genetic distance of Muslims from middle castes (F(ST) = 0.0090; DA = 0.0266) was significantly higher than that of Muslims from upper castes (F(ST) = 0.0050; DA = 0.0148)."
http://www.ncbi.nlm.nih.gov/pubmed/19130797
Parasar, i've been blogging about south asian genetics since 2002. your citations don't intimidate me. there's a lot of stuff out there. do you really expect me to be awesomely impressed by a 10 loci dendogram for example? (the main confound is that many north indian muslims do have ~5% west asian ancestry, which can cause spurious correlations because of the deep admixture of this among some upper castes).
do the basic math. even limiting to india there aren't enough brahmins/rajputs to account for how many muslims there are unless most of the brahmins and rajputs in india converted to islam. just like there aren't enough namboothiri brahmins to account for how many nasrani christians there are.
and yeah, i do know the backgrounds of some of these muslims. since i'm known in the genome blogger community people email me about stuff, and a few i recommended to HAP.
razib,
I just asked you a simple question - do you know where the HAP muslim samples are from. Rather than recite your blogging credentials, which I am familiar and sufficiently impressed with, please let me know the samples' backgrounds. Are they from UP or Bihar?
As to basic math: Do you know how many Brahmans and Rajputs there were before conversions in Bihar and UP (or the number of Nambuthiris in Kerala), as that number will need to be known to make a proper accounting?
Some conversions that come to mind:
Barahia Pathan Brahman/Rajput
Kamsar Pathan Brahman/Rajput
Zamania Pathan Brahman/Rajput
Bar Pathan Brahman/Rajput
Isa Afghan Bais Rajput
Salimpur Pathans, Baghocchia Brahmans, Bisen Rajputs of Majhauli and Saran.
As to the thought that W Asian ancestry and deep admixture is giving spurious signals - I doubt that looking at YDNA and mtDNA makeups of Muslims which is similar to that of Brahmans and Rajputs.
Y Samples mus 1-28 http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1380230/table/TB3/
mtDNA "modal haplotype in the pooled data set was also the modal haplotype in 34 (77%) of the 44 study populations. The 10 populations in which this haplotype was not the most frequent primarily comprised ethnic groups of the northern (Uttar Pradesh Brahmins, Punjab Brahmins, Rajput, Muslim) and the northeastern regions (Chakma, Jamatiya, Mog, Toto). In these populations, the modal haplotypes belong to non-M HGs, especially HG-U." http://genome.cshlp.org/content/13/10/2277.long
Are they from UP or Bihar?
there's one from UP fwiw. i don't want to get into an argument with you. some of your assertions are just funny IMO. i don't accept that the majority of the descendants of brahmisn and rajputs are muslims in south asia. that's the general inference if you take a robust view of your assertion.
razib,
"there's one from UP fwiw" Thanks for the info.
I was responding to this query - "why am i more similar to Bihari Birhaman than HRP0032 (Bihari Kayastha)?"
I would not have said the same about Bengal, for example, where the conversions were much more widespread and across the spectrum.
It makes more sense that the muslims converts were mostly from middle-low castes and of course you invariably will have some high caste converts as well. The conversion offers the lower castes a way to break out of the caste system so to speak and thus would be more attractive to the lower castes than the upper castes. Just my humble opinion.
Yes. I guess we don't have a representative Bihari Brahamin or Muslim genotype here, anyways. There might be a range and some Bihari Muslim might look like HRP0032 (Bihari Kayastha)
This leads to a question: How these reference C1, C2, etc are taken?
I use a large number of reference populations in my ADMIXTURE runs and the software computes these ancestral components C1, C2 etc based on allele frequencies.
Question : Whom does the SW Asian component peak among in your Reference III data-set? For whatsoever reason, my browser isn't able to sort by component.
Yemen Jews at 75%.
That's odd. I can't sort right now either. Was working yesterday.
Regarding the Yemeni Jews 75% SW Asian value. This may represent, for lack of a better term, a "pure" Arabian strain. Or, rather, one least (or less) affected by the significant population (and individual) movements, most noticeable beginning in the second half of the first millennium of the common era.
Something I posted on one of the forums recently.
David's ASD Matrix median values for Yemeni Jews and Yemeni(Behar et al dataset and 23andMe Assyrians):
Yemeni Jews
1 YEJ 0.24437
2 ASY 0.24820
3 CY 0.24830
4 SY 0.24838
5 KSA 0.24842
Yemeni
1 CY 0.25242
2 ASY 0.25265
3 GJ 0.25310
4 YEJ 0.25322
5 AM 0.25328
The "SW Asian" component is exceptionally uniform in most Harappa Middle Easterners:
Ethnicity "SW Asian" Total "African"
S Iraqi Arab 64% 10%
Iraqi Mandaean 58% 2%
IQ/Egy Jewish 61% 2%
Assyrian 58% 0%
Assyrian 59% 0%
Assyrian 58% 0%
And relevant "ref3" population SW Asian, individual African, and total African component values. Average values for the Harappa Assyrian participants included. Sorted in ascending order based on total African component value:
SW Asian European W African San/Pygmy E African Total African
ASY 58% 15% 0% 0% 0% 0%
AZJ 58% 16% 0% 0% 0% 0%
IQJ 62% 13% 0% 0% 1% 1%
IRJ 61% 13% 0% 0% 1% 1%
GEJ 60% 15% 0% 0% 2% 2%
SAM 69% 11% 0% 0% 2% 2%
DRZ 63% 15% 0% 0% 2% 2%
SYR 58% 15% 2% 0% 4% 6%
LEB 57% 16% 2% 0% 4% 7%
YEJ 75% 2% 0% 0% 8% 8%
PAL 62% 12% 3% 0% 6% 9%
KSA 70% 3% 2% 0% 7% 10%
JOR 58% 13% 4% 0% 7% 11%
BED 69% 6% 3% 0% 9% 13%
EGY 59% 8% 8% 1% 13% 21%
YEM 52% 4% 12% 2% 10% 25%
None of the populations with a predominantly Jewish or Christian tradition have detectable levels of the W African component based on this particular K11 (ref3) run.
Apologies. Jewish, Christian, and Druze traditions, I meant to say.
None of the populations with a predominantly Jewish or Christian tradition have detectable levels of the W African component based on this particular K11 (ref3) run.
Apologies. Jewish, Christian, and Druze traditions, I meant to say.
Paul, thanks for the percentages.
Also, none of them have any detectable San/Pygmy component (this is also true for non-Yemeni and non-Egyptian Arabs). Additionally, Christians and Azerbaijan Jews have no detectable African component. In fact, Christians of West Asia* are generally the genetically purest West Asians (I am calculating West Asian purity based on Caucasoidness), followed by Jews of West Asia* (including Samaritans), Druze and Mandaeans, and finally, by Muslims of West Asia, the genetically least pure West Asians.
* Yemen Jews are an exception, as they are in the same category in West Asian purity with Muslims of Asia.
* Yemen Jews are an exception, as they are in the same category in West Asian purity with Muslims of Asia.
A typo here. It would be "Muslims of West Asia" instead of "Muslims of Asia".
Zack, for some reason italics function isn't available in writing. So it isn't clear which parts of my last comments are quotation.
Fixed.
"In fact, Christians of West Asia*"
* is redundant here.
Simranjits,
That does not appear to be the case for UP at least. If you look at the Y-DNA data in the link above, UP Muslims are 0/19 H1. If UP Muslims were properly sampled, this data should be quite representative.
On the female side too, UP Muslims shared a feature seen in UP Brahmans, Punjab Brahmans, & Rajputs, that a particular M type was not modal.
This conversion of high castes can be explained by land-holding patterns. In the Akbar period over 90% of the landholders were Brahmans or Rajputs in the UP region (Ain-i-Akbari 1596AD*). Under the post Akbar regimes it became a lot easier to retain your holdings if you were a Muslim and this lead to a number of conversions.
*Eg. http://persian.packhum.org/persian/main?url=pf%3Ffile%3D00702052%26ct%3D114
75% of the Indian Muslim community is said to be Ajlaf or Arzal; no?
Zachary,
If Ashraf is defined as descendants of immigrants - Turk/Mongol, Arab, Persian - plus also includes Afghans and those promoted into the Ashraf, then 25/75% Ashraf/Ajlaf is quite likely.
HRP0095 and HRP0100 thought they had possible South Asian ancestry. That seems fairly unlikely at least in the last few generations since their Onge component is zero or very low.
Does that mean they don't qualify for participation in the project?
They are already in the project. 🙂
And even if they had a little bit of South Asian ancestry, they wouldn't form part of the core set which I use for South Asia only runs.
My question was : Will they be kicked out of the project now :p? Just joking of course..