View Full Version : South Asian DNA Tribes Results

08-02-2017, 10:55 AM
Anyone willing to share (post) their DNA Tribes Admixture Results? Mine are below (derived from AncestryDNA raw data).

I'm Caribbean Indian (second generation Indo-Trinidadian). All four grandparents are South Asian. Not sure if they're all from the same region though.

Continental Core Mixture
- South Asian 83.3%
- Asian 5.1%
- Siberian 4.7%
- North African and Middle Eastern 2.8%
- European 2%
- Pacific 1.6%
- Amerindian 0.4%
- Sub-Saharan African 0.1%

Regional Cluster Mixture
- Gangetic 32.6%
- Balochi Punjab 22.5%
- Dravidian 17.9%
- Orissan 11.9%
- Central Asian 9.6%
- Tajik Pashtun 2.7%
- Northeast Indian 1.8%
- Other 1%

Population Mixture
- Hindi Speakers in Uttar Pradesh India 28.9%
- Hindi Speakers in Haryana India 21%
- Marathi Speakers in Maharastha India 20.1%
- Dhurwa India 13.3%
- Karakalpaks 7.3%
- Spiti Speakers in Kashmir India 7.2%
- Wa 1.6%
- Mala Andhra Pradesh India 0.5%
- Other 0.1%

Population Matches: South Asia (Top 20)
Rank Population Map Area
- 1 Kol India
- 2 Chamar India
- 3 Muslim Uttar Pradesh India
- 4 Scheduled Caste Uttar Pradesh India
- 6 Tharu India
- 7 Kshatriya Uttar Pradesh India
- 8 Chenchu India
- 9 Gujarat India
- 10 Piramalai Kallar India
- 11 Mala Andhra Pradesh India
- 12 Bengali India
- 13 Kurumba India
- 14 Dharkar India
- 16 Gond India
- 17 Kurmi India
- 18 Lambadi India
- 19 Dusadh India
- 20 Scheduled Caste Tamil Nadu India

08-03-2017, 04:06 AM
Mine are here :)


08-03-2017, 08:46 AM
@ssamlal cool results. I'm curious what I would get. I feel like I would have roughly similar components to yours since on the wegene I got similar results to you. You might be from eastern UP or bihar since thats where most of the carribean indians came from and based on your genes I think its true.

@MonkeyDLuffy your results are interesting as well. You cluster halfway between caucasus and south india. Would you say its fairly accurate to your known lineage?

08-03-2017, 02:33 PM
I take DNA tribes results with grain of salt. They test very low SNPS, lower than majority of GEDMATCH tests. My closest population shows up as Sindhi, while I'm Punjabi. On the topic me being halfway between Caucasus and South India, it kind of makes sense for a North Western Indian (or Pakistani). I always score or show Central Asian in my results somehow, so who knows I might have some Turkic blood somewhere.

08-03-2017, 07:27 PM
Ok so here's my mixture. Just got my results back.

Continental Core Mixture
South Asian 90.2%
Asian 3.3%
North African and Middle Eastern 3%
Sub-Saharan African 1.6%
Pacific 1.4%
European 0.6%
Amerindian 0.1%

Regional Cluster Mixture
Gangetic 39.2%
Dravidian 23.9%
Balochi Punjab 20.1%
Pakistani 8.2%
Orissan 5.2%
Northeast Indian 1.6%
Nilotic 1.4%
Other 0.4%

Population Mixture
Sakilli India 33.4%
Hindi Speakers in Uttar Pradesh India 25.4%
Hindi Speakers in Haryana India 19.7%
Mala Andhra Pradesh India 10.3%
Brahui Pakistan 6.6%
Wa 1.7%
Anuak Ethiopia 1.4%
Hakkipikki India 0.8%
Bhunjia India 0.4%
Marathi Speakers in Maharastha India 0.2%
Other 0.1%

Population Matches:
Rank Population MapArea
1 Chamar India
2 Muslim Uttar Pradesh India
3 Kol India
4 Scheduled Caste Uttar Pradesh India
6 Kshatriya Uttar Pradesh India
7 Velama India
8 Tharu India
9 Gujarat India
10 Mala Andhra Pradesh India
11 Scheduled Caste Tamil Nadu India
12 Chenchu India
13 Dharkar India
14 Bhili Speakers in Rajasthan India
15 Lambadi India
16 Madiga Andhra Pradesh India
17 Kurumba India
18 Brahmin Tamil Nadu India
19 Hakkipikki India
21 Piramalai Kallar India
22 Hindi Speakers in Rajasthan India
23 Dusadh India
24 Meghawal India
25 Brahmin Andhra Pradesh India

#5 - Indian singaporean
#20- Cochin Jewish
both these populations were under diaspora.

Some of it is on point like the overall continental mixture. 90% south asian for me sounds perfect, but the regional clusters and everything beyond that seem slightly weird. I'm surprised I got more gangetic plain than OP given that he's from Uttar Pradesh/Bihar region.

Also a little confused on the 8.2% pakistani. What ethnicities does that include? Baloch punjab is a separate category, so I'm assuming like sindhi, brahui or kalash possibly? Not sure exactly. Some of these categories are interesting.

Also in the population matches, I didnt include it in the list above but I matched with Sindhi at #33 which was closer than #39-hindi speakers in Uttar pradesh, #40-brahmin uttarachal, #44- brahmins uttar pradesh, and #46-Bangladesh. I found this completely shocking since on most gedmatch calculators I matched way closer to those 4 populations than to sindhi.

Here's my cluster.

I'm right in the middle of gangetic cluster, being roughly equidistant to both balochi punjabi and dravidian, slightly closer to dravidian though.

@ssamlal Hey could you post your population cluster plot? I'm curious where you stand on that. I think you can't attach photos or post links since you're new on the forum. I had the same problem when i joined. Just upload your plot onto imgur.com, and then post the link onto this thread.

08-04-2017, 12:24 AM
Kush - Thanks for posting. Interesting that we have similar Wegene results but different DNA Tribes results.

Comparing our "Continental Core Mixture" it looks like I have lower South Asian and higher Asian / Siberian.

Comparing our "Regional Cluster Mixture" it looks like I have lower Gangetic / Dravidian and higher Orissan / Central Asian / Tajik Pashtun.

I also found it interesting to see high values for Marathi Speakers in Maharastha India & Karakalpaks in the "Population Mixture", assuming that my grandparents were from Bihar (or UP). This may just be my own lack of understanding. MonkeyDLuffy may be right to say take the results with a grain of salt.

Here is my MDS Cluster:

08-04-2017, 02:17 AM
Probably because Wegenes is more accurate as it tests more SNPs than DNA tribes

08-04-2017, 03:18 AM

08-04-2017, 03:29 AM

17917 1791817919

02-05-2018, 12:46 PM
Sorry for the bump from the abyss but I just thought khana's population match list looked awful. Sindhi makes sense I guess but the following populations aren't really a good fit. I guess I shouldn't judge too harshly since Indian Punjab is 4th for me.