Evidence of South Asian admixture in some Philipinos

[Tsakhur]

Pretty typical, maybe like a thinner, older Toshiro Mifune (with mustache/goatee)
Since his family is from the Metro Manila area, I think the odds of foreign contact over the centuries are both a lot higher and harder to nail down.
The Manila galleons alone sailed to/from there for 250 years.

I see. So if one would look at him, he/she wouldn't think he doesn't have West Eurasian ancestry? Does the rest of his family show European or Mediterranean features in their phenotypes or do they also look pretty typical Pinoy?

Interesting. Does his other Philippine DNA matches also have a lot of European?

[Kulin]

Agreed. It's pretty hilarious how a lot of Filipinos expected that they would have high West Eurasian due to Spaniard ancestors and then get really disappointed when it turned out to that they have like less than 5% or even close to zero in some cases. Most Pinoy gedmatch results seem to have around 1-5% West Eurasian range. More than 5% existed but as not common, also Filos who are 10% West Eurasian or more seem to be a minority, and those up to 15 to 20% Iberian seem to be very atypical or even rare.

But all of the 15 individuals do seem to have actual Indian ancestry just by looking at the fact that they also score minor Baloch? Or Baloch component is not enough to indicate that they have actual Indian gene flow?

I see. Yeah it could be precolonial or genetic variance. Btw, the first Filipina individual seem to be originally from Zamboanga which is in Mindanao, in Southern Philippines, closer to Indonesia and Malaysia, so maybe higher South Asian admixture there compared to the rest of the archipelago. Not sure about the second Filipina. Also I heard that there were recent Sepoy settlers in Cainta which is in Luzon in the 18th century and intermarried with locals along with recent Indian (mainly Punjabi) businessmen but Im not sure if it is too recent, for their genes to be absorbed into the general genetic pool of Pinoys.

Off topic but who do you think has more South Asian admixture in the region: Rakhine or Acehnese?

I think I would trust the g25 results more so than the admixture calculator ones. Certain noise amounts can be scored by most populations, but it does suggest some sort of ancestry in this case. They all seem to have Indian ancestry, but it isn't fully quantifiable with admixture calculators. The average Rakhine and Achehnese probably have around quite similar W eurasian, but I would think that some communities of the former have more. Also, Rakhines are patrilineally Tibeto-Burman-like but 2/3rds of their mtdna are of Indian/W Eurasian origin (i.e. common South Asian varieties of M and things like T2). They had a practice of pirate raiding in Bengal, and the word 'mog' (from Magadhi) in Bengali means Arakanese but also has a connotation of pirate. Even the daughters of the Mughal prince, Shah Shuja were also enslaved and taken as queens by the Kings there while on refuge.

[Tsakhur]

I think I would trust the g25 results more so than the admixture calculator ones. Certain noise amounts can be scored by most populations, but it does suggest some sort of ancestry in this case. They all seem to have Indian ancestry, but it isn't fully quantifiable with admixture calculators. The average Rakhine and Achehnese probably have around quite similar W eurasian, but I would think that some communities of the former have more. Also, Rakhines are patrilineally Tibeto-Burman-like but 2/3rds of their mtdna are of Indian/W Eurasian origin (i.e. common South Asian varieties of M and things like T2). They had a practice of pirate raiding in Bengal, and the word 'mog' (from Magadhi) in Bengali means Arakanese but also has a connotation of pirate. Even the daughters of the Mughal prince, Shah Shuja were also enslaved and taken as queens by the Kings there while on refuge.

Agreed. I also think that G25 would be more reliable than those of Gedmatch or other Admix calcs. Why do you think it's not fully quantifiable? Is it because some of the South Asian in them might actually be something Negrito or local AASI-like rather than actual Indian ancestry?

I would presumed that the Moros (Muslims) of Mindanao and other islands in the Southern part of the Philippine archipelago which is much closer to Indonesia and Malaysia would have higher Indian ancestry and maybe even Arab admixture than the majority of Philippines who are mostly Christian. But I am not sure if my hypothesis is true as I haven't seen any DNA study on the Muslim Pinos and the two Moro gedmatch results that I seen, score almost like the average Pinoy, with very low South Asian and very negligible to almost zero West Eurasian admixture. But its only two results, so I won't jump to a conclusion yet.

Very interesting. Thanks for the explaining and educating me on this. Is it possible that the Rakhine can be around 30-40% South Asian which is roughly 15-20% West Eurasian on average? It seems they and the Acehnese are one of the most West Eurasian admixed ethnic groups in SE Asia. Yes, it make sense that some communities of the Rakhine would possess more due to their prolong historical connections with Bengal. If the Rakhine and Acehnese can have approximately 15-20% West Eurasian, that would make them as Western admixed as the Maasai and other Cushitic admixed populations from East Africa. Btw do you think this Burmese individual is an ethnic Rakhine? I can easily modelled her as 40% South Asian in Dodecad K12b using Vahaduo and she is almost 20% West Eurasian in PuntDNAL K13 and other cals when I modelled her using Vahaduo?: https://anthrogenica.com/showthread....l=1#post724556

This is something that I observed about Indonesia based on the Gedmatch kits of users from there. It seems like Sumatra and its ethnic groups like the Acehnese, Minangkabau, Batak (I don't think I have seen a Batak gedmatch kit but I have seen their pop averages on MDLP K23b, and they seem to score like Malays and Cambodians in their South Asian score), Lampung, etc. likely has the most West Eurasian ancestry among the islands in the archipelago followed by Jakarta, possibly Western areas (ethnic Sundanese-dominated regions) of the Java island and then Bali. Sulawesi (I think I have seen only four gedmatch kits from there based on their names and South Asian score), Borneo and Eastern Indonesia like Sumba, Flores, Timor, Maluku seems to have very minor to negligible West Eurasian admixture by comparison. For some really strange reason, a lot of Indonesian gedmatch kits from Java especially ethnic Javanese (I know they are Javanese because of their given and lastnames) who are from Central and Eastern Java, despite scoring decent amount of South Asian components in many calculators, have very negligible amounts of West Eurasian ancestry and when they have West Eurasian, it tends to be something European (likely Dutch) or even West Asian-related rather than Indian-mediated Western ancestry which really perplexed me. Its the same case in G25 as well, for some reasons the Indonesian_Java samples have only 1-2.4% West Eurasian ancestry which is less than the Indonesian_Bali/Balinese samples who are around 3-5% West Eurasian. On the contrary, I have seen some Javanese gedmatch kits who have a decent amount of West Eurasian (5% or more) ancestry as well, but those who scores negligible West Eurasian (like 1-3%) for some strange factor, seems to be much more prevalent.

I thought that scoring a lot of South Asian component would also guarantee a nice amount of Iran_N and minor Steppe, but that seems not to be the case for many Indonesians. Could it be that the high "South Asian" that a lot of Indonesians from Java are scoring is something Onge, Negrito or local AASI-like ancestry rather than actual Indian gene flow?

Do you know what could be the plausible factors behind this? I would have expected that the Javanese would be more West Eurasian like the Khmers and other SE Asians, considering how much Indianized they are culturally during pre-islamic times to even today (looking at their first and lastnames for example). Is it possible that Indian traders or settlers only settled mainly in urban and coastal areas but not in the more remote or rural areas of Java hence very little gene flow in a lot of Javanese results (both Gedmatch and G25 samples)? I can post you some Javanese results to demonstrate what I have observed. Could it be that, due to Bali being much smaller, it is easier for Indians historically
to settle down and intermarried with the locals, thus introducing their ancestry into the local Balinese gene pool than in Java.

Btw I have a lot of SE Asian gedmatch kits and I really wanted to email and ask them to try out G25, but the thing is I'm too shy to do it since I haven't done any DNA test yet so I might not even be their relatives. Do you mind persuading them for me please if I give you their gedmatch kits? (I collected lots of them)

[Kulin]

Agreed. I also think that G25 would be more reliable than those of Gedmatch or other Admix calcs. Why do you think it's not fully quantifiable? Is it because some of the South Asian in them might actually be something Negrito or local AASI-like rather than actual Indian ancestry?

I would presumed that the Moros (Muslims) of Mindanao and other islands in the Southern part of the Philippine archipelago which is much closer to Indonesia and Malaysia would have higher Indian ancestry and maybe even Arab admixture than the majority of Philippines who are mostly Christian. But I am not sure if my hypothesis is true as I haven't seen any DNA study on the Muslim Pinos and the two Moro gedmatch results that I seen, score almost like the average Pinoy, with very low South Asian and very negligible to almost zero West Eurasian admixture. But its only two results, so I won't jump to a conclusion yet.

Very interesting. Thanks for the explaining and educating me on this. Is it possible that the Rakhine can be around 30-40% South Asian which is roughly 15-20% West Eurasian on average? It seems they and the Acehnese are one of the most West Eurasian admixed ethnic groups in SE Asia. Yes, it make sense that some communities of the Rakhine would possess more due to their prolong historical connections with Bengal. If the Rakhine and Acehnese can have approximately 15-20% West Eurasian, that would make them as Western admixed as the Maasai and other Cushitic admixed populations from East Africa. Btw do you think this Burmese individual is an ethnic Rakhine? I can easily modelled her as 40% South Asian in Dodecad K12b using Vahaduo and she is almost 20% West Eurasian in PuntDNAL K13 and other cals when I modelled her using Vahaduo?: https://anthrogenica.com/showthread....l=1#post724556

This is something that I observed about Indonesia based on the Gedmatch kits of users from there. It seems like Sumatra and its ethnic groups like the Acehnese, Minangkabau, Batak (I don't think I have seen a Batak gedmatch kit but I have seen their pop averages on MDLP K23b, and they seem to score like Malays and Cambodians in their South Asian score), Lampung, etc. likely has the most West Eurasian ancestry among the islands in the archipelago followed by Jakarta, possibly Western areas (ethnic Sundanese-dominated regions) of the Java island and then Bali. Sulawesi (I think I have seen only four gedmatch kits from there based on their names and South Asian score), Borneo and Eastern Indonesia like Sumba, Flores, Timor, Maluku seems to have very minor to negligible West Eurasian admixture by comparison. For some really strange reason, a lot of Indonesian gedmatch kits from Java especially ethnic Javanese (I know they are Javanese because of their given and lastnames) who are from Central and Eastern Java, despite scoring decent amount of South Asian components in many calculators, have very negligible amounts of West Eurasian ancestry and when they have West Eurasian, it tends to be something European (likely Dutch) or even West Asian-related rather than Indian-mediated Western ancestry which really perplexed me. Its the same case in G25 as well, for some reasons the Indonesian_Java samples have only 1-2.4% West Eurasian ancestry which is less than the Indonesian_Bali/Balinese samples who are around 3-5% West Eurasian. On the contrary, I have seen some Javanese gedmatch kits who have a decent amount of West Eurasian (5% or more) ancestry as well, but those who scores negligible West Eurasian (like 1-3%) for some strange factor, seems to be much more prevalent.

I thought that scoring a lot of South Asian component would also guarantee a nice amount of Iran_N and minor Steppe, but that seems not to be the case for many Indonesians. Could it be that the high "South Asian" that a lot of Indonesians from Java are scoring is something Onge, Negrito or local AASI-like ancestry rather than actual Indian gene flow?

Do you know what could be the plausible factors behind this? I would have expected that the Javanese would be more West Eurasian like the Khmers and other SE Asians, considering how much Indianized they are culturally during pre-islamic times to even today (looking at their first and lastnames for example). Is it possible that Indian traders or settlers only settled mainly in urban and coastal areas but not in the more remote or rural areas of Java hence very little gene flow in a lot of Javanese results (both Gedmatch and G25 samples)? I can post you some Javanese results to demonstrate what I have observed. Could it be that, due to Bali being much smaller, it is easier for Indians historically
to settle down and intermarried with the locals, thus introducing their ancestry into the local Balinese gene pool than in Java.

Btw I have a lot of SE Asian gedmatch kits and I really wanted to email and ask them to try out G25, but the thing is I'm too shy to do it since I haven't done any DNA test yet so I might not even be their relatives. Do you mind persuading them for me please if I give you their gedmatch kits? (I collected lots of them)

'Indianization' in the phillipines probably followed a South-North axis. That is the reason why people in Mindanao and the Sulu archipelago remained culturally "malayic" vs hispanicized/christianized in much of the country. Some groups in the phillipines (like in Indonesia, Malaysia etc), especially in the north were also never exposed to this Indian influence, and remained a more secluded animist lifestyle.

Most SE Asians likely have ancestry from South India, and Javanese scoring like that may be a reflection of that, or more likely discrepancy from admixture calculators IMO. Also, vast majority of Gedmatch members won't reply, and a lot are protective about their data. Most testers do not make it the "gedmatch phase" and are okay altogether with 23andme, myheritage etc. A lot of people don't even connect on 23andme itself lol. I think if they had the desire to go for further testing, they'd likely find the will to do so independently or even come across forums such as this.

The Rakhine are definitely atleast 30% "Bengali-like" as are neighbouring groups (Tripuri, Chakma etc). I've been to the Chittagong hill tracts, where a lot of Rakhine live, and it isn't possible to distinguish the Rakhine from Chakma and Tripuri for the most part, though Tripuri people are usually darker. Coastal Rakhines are also probably more W Eurasian than the hill ones. Most don't look South Asian, but the influence is visible in a lot.

[Tsakhur]

'Indianization' in the phillipines probably followed a South-North axis. That is the reason why people in Mindanao and the Sulu archipelago remained culturally "malayic" vs hispanicized/christianized in much of the country. Some groups in the phillipines (like in Indonesia, Malaysia etc), especially in the north were also never exposed to this Indian influence, and remained a more secluded animist lifestyle.

Most SE Asians likely have ancestry from South India, and Javanese scoring like that may be a reflection of that, or more likely discrepancy from admixture calculators IMO. Also, vast majority of Gedmatch members won't reply, and a lot are protective about their data. Most testers do not make it the "gedmatch phase" and are okay altogether with 23andme, myheritage etc. A lot of people don't even connect on 23andme itself lol. I think if they had the desire to go for further testing, they'd likely find the will to do so independently or even come across forums such as this.

The Rakhine are definitely atleast 30% "Bengali-like" as are neighbouring groups (Tripuri, Chakma etc). I've been to the Chittagong hill tracts, where a lot of Rakhine live, and it isn't possible to distinguish the Rakhine from Chakma and Tripuri for the most part, though Tripuri people are usually darker. Coastal Rakhines are also probably more W Eurasian than the hill ones. Most don't look South Asian, but the influence is visible in a lot.

Do you think Pinoys from Mindanao or Sulu islands will have more Indian or even Arab ancestry compared to the rest of the country?

Would you say most SE Asians have West Eurasian ancestry or are those who have very little to none like Vietnamese, Filipinos are more common? Can you elaborated more on the parted that I bolded in your sentence " Javanese scoring like that may be a reflection of that, or more likely discrepancy from admixture calculators"? I still don't understand.

That really sucks. You are right. I have tried to persuade some 23andme users on Reddit to go further and do Gedmatch, and they seem to ignore or didn't even read my comments. It seems most 23andme reddit users don't even know about the existence of Gedmatch or anything else besides 23andme or other commercial DNA testing companies. By not connecting with 23andme, do you mean they don't understand or can't accept their DNA results? Looks like persuading most of them won't work at all.

Damn, I really wanted David to upload more SE Asian samples into G25. Maybe I will have to find some papers for him to do that. It would be great if someone can create a Gedmatch calculator that also have the AASI along with other West Eurasian components, but I feel it will never happened. I think the closest one was Eurasia K9 ASI but he removed the calc, after he got banned in this forum. Is there a way to have Eurasia K9 ASI included back on Gedmatch or Vahaduo?

Have you personally sucessfully persuaded any South Asian gedmatch user to try G25 or as you mentioned, most of them never replied?

Do you have some Rakhine gedmatch results or samples that can be uploaded to G25 by David? Its interesting to hear that the Tripuri and Chakma are also approximately 30% Bengali-like. Do the Rakhine and Tripuri speak similar languages?

[Kulin]

Do you think Pinoys from Mindanao or Sulu islands will have more Indian or even Arab ancestry compared to the rest of the country?

Would you say most SE Asians have West Eurasian ancestry or are those who have very little to none like Vietnamese, Filipinos are more common? Can you elaborated more on the parted that I bolded in your sentence " Javanese scoring like that may be a reflection of that, or more likely discrepancy from admixture calculators"? I still don't understand.

Personally, I think overwhelming majority of South-east Asians (groups that have been Hindu in the past) have some South Asian ancestry. I meant that Javanese not showing much 'w eurasian' input is likely a fault of admixture calculators. I would wager they would show a similar amount as some others using g25. Groups devoid of South Asian ancestry are ones that were under Chinese influence (e.g. Vietnamese), or groups that never were in the Hindu-Buddhist cultural sphere (a big percent of the Phillipines + isolated groups in Borneo or Melanesians etc).

That really sucks. You are right. I have tried to persuade some 23andme users on Reddit to go further and do Gedmatch, and they seem to ignore or didn't even read my comments. It seems most 23andme reddit users don't even know about the existence of Gedmatch or anything else besides 23andme or other commercial DNA testing companies. By not connecting with 23andme, do you mean they don't understand or can't accept their DNA results? Looks like persuading most of them won't work at all.

Yep, majority of people I've sent invites to in 23andme do not accept it, which is unfortunate. Though if I did not know much about these things, I would be wary of sending data to a stranger as well, so it is understandable to some degree.

Damn, I really wanted David to upload more SE Asian samples into G25. Maybe I will have to find some papers for him to do that. It would be great if someone can create a Gedmatch calculator that also have the AASI along with other West Eurasian components, but I feel it will never happened. I think the closest one was Eurasia K9 ASI but he removed the calc, after he got banned in this forum. Is there a way to have Eurasia K9 ASI included back on Gedmatch or Vahaduo?

Have you personally sucessfully persuaded any South Asian gedmatch user to try G25 or as you mentioned, most of them never replied?

Do you have some Rakhine gedmatch results or samples that can be uploaded to G25 by David? Its interesting to hear that the Tripuri and Chakma are also approximately 30% Bengali-like. Do the Rakhine and Tripuri speak similar languages?

I wouldn't trust Lukasz's calculators to be fair, I'm not sure what algorithm it uses so I can't comment on that. I have persuaded people that post here, but not anyone from outside this forum unfortunately.

I have seen some Rakhine results, but not that many. Also, do not really know anyone who tested from that background, especially since I quit Gedmatch due to their privacy issues. Having said that, to give you an estimate, Tripuris are on G25 and their fst value to even the Bamar groups are around 3, which is pretty close (closer than I am for e.g. to the g25's Bengali average). This would mean that Tripuris and the Rakhine are especially genetically close, since they have always had close interactions throughout history.
Tripuris speak kokborok, a Bodoish language, while Rakhine basically speak a dialect of Burmese. Rakhine language mainly just pronounces words differently (r instead of y for e.g. Rangon instead of Yangon, Mranma instead of Myanma) and has more sanskrit/persian loanwards. Both are of course Tibeto-Burman in origin.

[Tsakhur]

Personally, I think overwhelming majority of South-east Asians (groups that have been Hindu in the past) have some South Asian ancestry. I meant that Javanese not showing much 'w eurasian' input is likely a fault of admixture calculators. I would wager they would show a similar amount as some others using g25. Groups devoid of South Asian ancestry are ones that were under Chinese influence (e.g. Vietnamese), or groups that never were in the Hindu-Buddhist cultural sphere (a big percent of the Phillipines + isolated groups in Borneo or Melanesians etc).


Yep, majority of people I've sent invites to in 23andme do not accept it, which is unfortunate. Though if I did not know much about these things, I would be wary of sending data to a stranger as well, so it is understandable to some degree.


I wouldn't trust Lukasz's calculators to be fair, I'm not sure what algorithm it uses so I can't comment on that. I have persuaded people that post here, but not anyone from outside this forum unfortunately.

I have seen some Rakhine results, but not that many. Also, do not really know anyone who tested from that background, especially since I quit Gedmatch due to their privacy issues. Having said that, to give you an estimate, Tripuris are on G25 and their fst value to even the Bamar groups are around 3, which is pretty close (closer than I am for e.g. to the g25's Bengali average). This would mean that Tripuris and the Rakhine are especially genetically close, since they have always had close interactions throughout history.
Tripuris speak kokborok, a Bodoish language, while Rakhine basically speak a dialect of Burmese. Rakhine language mainly just pronounces words differently (r instead of y for e.g. Rangon instead of Yangon, Mranma instead of Myanma) and has more sanskrit/persian loanwards. Both are of course Tibeto-Burman in origin.

Agreed. I also believe that the vast majority of SE Asians have South Asian ancestry. Thanks for explaining further. That's the reason I wished I can asked those Javanese gedmatch kit users to buy G25 so it can solve once for all whether their West Eurasian is hiding under the South Asian components that they are scoring as a result of admixture calc limitations. I see. I also believed that they should have almost the same percentage of West as the Indonesia_Java samples or other SE Asians on G25. Concurred regarding the groups that were never Indicized who seem to be rather the minority in the region.

I really wished there are more people who are genuinely interested in learning more about their DNA and just the subject of genetics in general, which seems to be a small minority among the DNA testers.

Point taken about Lukasz's calcs. But Eurasia K9 ASI was created by Kurd, who after being banned from Anthrogenica, removed most of his calculators from Gedmatch, probably as a result of outrage and anger at being banned. Kurd might have an agenda, but I feel the forum shouldn't banned him. Because it affects those including me who thinks his calcs might be useful. For those who you persuaded in this forum, does it work out?

Were these Rakhine results from Gedmatch, 23andme or AncestryDNA? You don't have your Gedmatch account any longer? Damn, the website seems ok now though.

Ah ok. Rakhine and Tripuris would be very similar including the South Asian and West Eurasian then. It would still be great to have Rakhine G25 samples because I suspected they would be more Western-shifted than even the Bamars. It really sucks that there are only 3 Burmese samples on G25. And all of them are also ethnic Bamar? I feel we need much more Burmese samples including from other ethnic groups like Mon, Shan, Kayin, Rakhine, Kachin, Kayah, etc. to gain a better and clearer genetic insight of Burma.

Btw off topic, but do you have some Khasi samples that can be uploaded to G25? I suspected that they would be around 13-15 or more% West Eurasian, with the rest of their genomes primarily East Asian+very high AASI?

[cmd06]

Aren't most filipinos closely related to malay people like indonesian and malaysian? Why is that almost all gedmatch result of Filipinos, the single population source is almost always Kinh/Vietnamese? Even the secondary population (source) distance is almost always Kinh/Vietnamese + some other population.

The closest malay combination that I saw was 60% singaporean malay + singaporean chinese, sometimes even 50-50.

[Tsakhur]

Aren't most filipinos closely related to malay people like indonesian and malaysian? Why is that almost all gedmatch result of Filipinos, the single population source is almost always Kinh/Vietnamese? Even the secondary population (source) distance is almost always Kinh/Vietnamese + some other population.

The closest malay combination that I saw was 60% singaporean malay + singaporean chinese, sometimes even 50-50.

I believe its because they are much more "purer" Austronesian genetically than Indonesians and Malays who have substantial South Asian or AASI ancestry. And it might have to do the fact that Indonesians and Malays also have a lot of Austroasiatic/mainland SE Asian-like ancestry than Pinoys mostly lack.

[cmd06]

But viets are austroasiatics right? Or their austroasiatic component did not come from the viets but rather from another population like cambodians? Indonesians/malaysians usually have cambodia as their single population source then modeled as malay + some other population.