I have been reporting only individual admixture results for Harappa Project participants. I think it's way past time I posted some group averages too.
You can see the groups I have assigned participants and the current count for each group.
The average admixture results for each group are in a spreadsheet. This is using Reference 3. You can compare with the reference population results.
Here's the bar chart for participants group averages. Remember you can click on the legend or the table headers to sort.
The Haryana Jatt having a high concentration of European is really interesting, why do you think this is?
It's hard to say. All Jatts seem to have higher European. That can be explained by may be partly foreign origin. But the higher numbers for Haryana Jatts compared to Punjabi Jatts are a mystery.
Interesting thing is that the more South of the Indian Sub Continent you go the lower the European concentration. Could be the extent of Aryan migration just concentrated in the North?
Zack, what are your criteria for assignment in the averaged groups? All four grandparents from the same group? Also, no relatives are included in the group averages, right?
Yes and yes.
In the individual ethnicities list there seem to be 3 Romanies with all four grandparents Romany, whereas in the above Romany group there are 4 Romanies. Does one of them have at least one non-Romany grandparent? Or, is one of the full Romanies non-identified in the individual ethnicities list?
I had one of the partial Romany in there.
Zack, could you add a list which shows which individuals you inluded in the averaged groups? Or alternatively, you can add population portaits of the averaged groups together with the individual IDs.
Already Included it in the post.
I think you are making a big mistake here, other than participant no. 21 in your database no one is racially kashmiri here, the other one or two guys who are presenting themselves as kashmiri here are actually pothoharis of so-called Azad kashmir who are racially NOT kashmiri, I find it offensive to consider someone as kashmiri when he does not originate from the valley of kashmir and does not speak kashmiri/Kashur language. Please rectify this mistake.
Thanks for posting these averages, Zack.
The Karnataka Iyengars of the project actually speak a dialect of Tamil called Hebbar Iyengar Tamil. So, they're specifically like Tamil Brahmins who reside in Karnataka. Iyengars can be found in all the southern states with the exception of Kerala, although obviously being concentrated in Tamil Nadu. Likewise, Iyers too can be found in all southern states with the exception of Andhra Pradesh. So, I'm not too sure how accurate it is to include the Karnataka Iyengars in the Karnataka Brahmin average.
Yes, I am thinking of splitting up the Iyers and Iyengars and keeping the other South Indian Brahmins separate.
Great. Also, Dodecad's Iyengar_D is identical to the participants of the same group you have in your project. Of course, that particular population portrait is devoid of any relatives.
OK I have split the Iyers and Iyengars, though no one would know the difference since their admixture proportions are so similar. š
Also, I don't include the relatives (i.e., nonfounders) in computing the population averages.
Sorry, actually one of your Iyengars is actually not in Dodecad. Well, almost identical then :-).
".. though no one would know the difference since their admixture proportions are so similar."
True that :)! We're a boringly similar bunch indeed.
Already Included it in the post.
Is it? I meant a list which only includes the individuals you used in the calculation of the group averages together with the name of the group they are a part of.
Fascinating and informative, hope more groups will be added.
Interesting is that groups like Georgian and Turkish are also showing quite high South asian composition! My question is:
1. Do the participants of the two groups have any relation individually to the subcontinent?(if you know).
2. If not then what is the explanation?
Have a good time.
I wrote my last comment in the wrong place, so I am rewriting it:
Already Included it in the post.
Is it? I meant a list which only includes the individuals you used in the calculation of the group averages together with the name of the group they are a part of.
This spreadsheet tells us the number of participants in each group + average admix proportions in each group - https://docs.google.com/spreadsheet/ccc?key=0AuW3R0Ys-P4HdENxSm4xSTAtcnZwcWFOMVJSdHJjM3c#gid=17
This the overall count of individuals of specific groups - https://docs.google.com/spreadsheet/ccc?key=0AuW3R0Ys-P4HdGE4eDh6emt1dUs2U2pXTkVjS0lsV1E#gid=2
And this, individual ID, ethnicity, the group they belong to in these averages calculated and their uniparental haplogroups - https://docs.google.com/spreadsheet/ccc?key=0AuW3R0Ys-P4HdGE4eDh6emt1dUs2U2pXTkVjS0lsV1E#gid=0
Those spreadsheets do not show exactly which individuals were used in the calculation of group averages.
Dude. Look a the Group column of the last spreadsheet. Whatever group a given ID belongs to, was an ID used to calculate the averages posted by Zack.
There are many multi-ethnic individuals in that spreadsheet, some of them may have been used in the group averages (Zack already said that he used a mixed Romani in the Romani average), and it is not clear which were used and which were not.
I have no idea why you can't read a simple spreadsheet. The ID column shows the IDs of all participants. The Ethnicity column gives the detailed ethnicity info for everyone. And the Group column lists which group average that participant is included in. Blank means not included in a group.
Most of the multi-ethnic individuals haven't been assigned to a group.
I think you have mentioned in one of your posts that participant no. 21 is your wife and is kashmiri pandit, I would like to draw your attention towards a very important fact that the participant no. 44 who is presenting himself here as kashmiri is actually a pothohari/punjabi speaking person from Azad Jammu/kashmir area administered by pakistan. So the only participant in this database who is racially kashmiri, that is originating from the Valley of kashmir is participant no. 21. It is really strange that the administrator of this site is totally unaware of the ground realities of kashmir region, I would like that the kashmir cluster should consist of only participant no. 21 who is the sole racially kashmiri participant in this database.
Interesting is that groups like Georgian and Turkish are also showing quite high South asian composition! My question is:
1. Do the participants of the two groups have any relation individually to the subcontinent?(if you know).
2. If not then what is the explanation?
Have a good time.
Hgh South Asian composition?! If you mean the "South Asian" component, their "South Asian" component percentages are not particularly high for their region. Much more importantly, the South Asian" component is not a South Asia or South Asianness-specific component, it is clusters with the "European" and "SW Asian" components, so it is a West Eurasian component rather than South Asian.
Sorry for the typos. I had a hurry while writing that post. Here is a typo-free version:
High South Asian composition?! If you mean the "South Asian" component, their "South Asian" component percentages are not particularly high for their region. Much more importantly, the "South Asian" component is not a South Asia or South Asianness-specific component, it clusters with the "European" and "SW Asian" components, so it is a West Eurasian component rather than South Asian.
I have no idea why you can't read a simple spreadsheet. The ID column shows the IDs of all participants. The Ethnicity column gives the detailed ethnicity info for everyone. And the Group column lists which group average that participant is included in. Blank means not included in a group.
Sorry, I did not see the group column. Somehow I scrolled too much right and left and missed that one column.
Funny that the half Romany half, Serb individual is genetically more Romany than all the full Romany participants (all of them from England). It seems the ancestors of English Romanies mixed too much with Europeans on their way to England. It teaches us not to overgeneralize about geographically widespread ethnic groups like Jews and Romanies.
It seems the ancestors of English Romanies mixed too much with Europeans on their way to England.
including the mixing after the arrival to England
If The South Asian component clusters with SWA and European components then,
1. Is The component is Ancestral north indian one? Grouped together as the South Asian component?
2. If it is "West eurasian" then why it have to be named after South Asian?
Another thing is that metsupalu et al. found ANI component to be atleast 12kya in south asia and it to be incompatible and older for the "Obvious" explanation.Were they also Indigenous Aryan fanatics?
Have a good time.
Nirjhar,
I am not talking about where the components originated, that is a too speculative subject. The "South Asian" component is most closest to the "European" and "SW Asian" components and even clusters with them, that is why it is a West Eurasian component. It is certain that all these three West Eurasian components have a relatively recent common origin. The "South Asian" component does not have to be named "South Asian" or anything else, component names are just mnemonic tools. It is obvious that Zack named it "South Asian" just because that it peaks in South Asia. Why does it peak in South Asia? Because, as AV explained, the "South Asian" component includes a small ASI fraction when it occurs in South Asian populations but lacks it in West Eurasian populations.
Zack. Would it be possible to end the Reference 3 K=11 era by assessing Xing et al populations in terms of Ref 3 K=11? Dienekes was able to do the same with his Dodecad v3 using 13,000 SNPs (presumably for each population).
There's a bunch of reference 3 related stuff coming while I am working on the next version.
Okay thats very nice but the "relatively recent ancestor" is probably older than neolithic as for the metspalu et al.
and ofcourse the origin place it can be anywhere N. India, C. Steppes, Caucasian etc but no researcher have sofar tried to find that like metspalu et al. did to find the least age of the ANI specific component in S. Asia.
Have a good time.