PDA

View Full Version : East Asians, Southeast Asians, and Pacific Islanders and etc' Genetic Discussion



Timawa
08-23-2016, 06:25 AM
There was not a lot of discussion about Asians' genetic studies going here. I will post various charts and link them to articles.

http://i45.photobucket.com/albums/f87/redwine88/Mobile%20Uploads/5ADDE87E-F10F-49B0-867B-551607BCFF5A_zps8ltjsjzo.jpg

http://www.nature.com/articles/srep14375

http://i45.photobucket.com/albums/f87/redwine88/srep26066-f5_zpssi01tr1r.jpg

http://www.nature.com/articles/srep26066#f5

Iloko
08-23-2016, 08:18 AM
According to the study the pink bar is "Proto-Malay". Proto-Malays are a hybrid of mainland-Negrito + East Asian component right? What's the difference from 'Austro-Asiatic' if any?

Ingenious
08-23-2016, 06:11 PM
http://i.imgur.com/QFMRstu.png
http://i.imgur.com/jXCdYml.png

quite interesting that chinese have finnish

Timawa
08-23-2016, 08:07 PM
According to the study the pink bar is "Proto-Malay". Proto-Malays are a hybrid of mainland-Negrito + East Asian component right? What's the difference from 'Austro-Asiatic' if any?

Malaysians and Javanese are the hybrid of Austronesian and AustroAsiatic with a minor S. Indian or Negrito admixture. Viets, Thais, and Cambodians are mostly AustroAsiatic, I think.

Can you understand the PCA plot?
http://i45.photobucket.com/albums/f87/redwine88/srep26066-f5_zpssi01tr1r.jpg

Iloko
08-24-2016, 02:18 PM
Malaysians and Javanese are the hybrid of Austronesian and AustroAsiatic with a minor S. Indian or Negrito admixture. Viets, Thais, and Cambodians are mostly AustroAsiatic, I think.

Can you understand the PCA plot?
http://i45.photobucket.com/albums/f87/redwine88/srep26066-f5_zpssi01tr1r.jpg
To understand it we'll have to know exactly how PCA plots work!..Dr McDonald also does PCA plot analysis. He mentions it in this guide:


Interpreting the Results of the BGA Analysis

In this analysis of BGA (biogeographical ancestry) people get, as a minimum, four types of data, the first in text, and the other three presented as visual representations that you click on to display.
1) Quantitative Results: A list, shown in the body of the e-mail, of possible sets,
one set per line, of possible ancestry. Some people don't get this list if they test
as, within an error tolerance, 100% a certain ancestry. Each line represents
the computer algorithm’s guess as to the test person’s makeup. The most likely sets in a statistical sense are at the top. However, this does not take into account what is called “prior knowledge”. The classic example of this is with Afro-(Euro)Americans: the computer quite frequently places Hungary or Romania at the top of the list, and England and Ireland below these, sometimes even too low to appear on the list it automatically send to people. But we know that most such people have European ancestors from Western Europe, which is “prior knowledge”. This means that in reality, all things considered (and you really need to consider all the info you have) , Hungarian is less likely than English unless you suspect Hungarian. There are other such cases, especially relating to Jewish versus Italian or Mediterranean ancestry. I provide written comments in these cases. Note especially that we lack specific comparisons for Native American ancestry from the USA and Canada east of the Rocky Mountains; people with such ancestry, as well as people from the Caribbean and central Mexico (Aztec, etc.), come out as “Maya”.
2) Plot or Plots with Dots (“Scatter Plots”): These plots are labeled with comparison populations and a crosshair icon representing the testee. This is called the PCA (Principal Component Analysis) "scatter" plot. See below for how to read these. The computer generates several dozen of these, and I send only ones of interest for whatever makeup the test person is. There are a couple of special ones that test for specific ethnic ancestry. The most common of these is for Jewish, but others exist for such ethnicities as Sardinian, Kalash, Pygmy, and certain Native American tribes. None exist, unfortunately, for East Asian ethnic people.
3) Chromosome Segment Painting: This feature gives a “painting” of the testee’s
22 autosomes and (unless your data is from FTDNA and you
did not send the X-specific file) the X chromosome which are "color coded"
to show the "continental" ancestry found at places along them. This test is very important for people with small amounts of ancestry from a “continent” differing from their major ones. It can detect, just for example, as small as 0.4% Native American in a pure European or a mixed Afro-(Euro)American person even if the test described in #1 above cannot. Such results will be mentioned in the text. In the specific case mentioned above the American comes out of the European part.
4) "Spot on the Map Plot": Here the data (as computed from number one above) is plotted as a summary. The green spot represent the average location of all ancestry except Native American and Oceanic. Note that for people with significant sub-Saharan African ancestry this green spot is essentially meaningless. But for people with just European, Mideast, and sometimes South Asian ancestry it is usually quite accurate as far as their average ancestry goes. The geographic map plot shows each continent’s average location, as well as
an overall average (green) excluding the American and Oceanic ancestry.
The “chromosomes” plot just shows which geographic region each part of each chromosome is most likely to have come from. Since everyone has two copies of each chromosome (except males for the X), they may be split in half. But note that while distant regions such as Europe and Africa split cleanly, nearby ones like Europe and the Mideast don’t. As a result often southern Europeans will show some Mideastern spots without having recent ancestry from the Mideast. Mideasteners, especially from North Africa often have sub-Saharan African regions, as do Mediterranean Europeans. Small regions (< 2 Mb) are often just “noise. I have special tests that can sometimes tell whether they are or not, and that will reported in the text if significant. Small regions may not show on the “quantitative” lists and will also be discussed if significant. It is important to remember that this plot cannot tell father from mother in any way.
The PCA “scatter” plots are the result of a calculation across the whole genome (all the chromosomes except the X) and report the average ancestry. If a person’s ancestry is from one place that will be reflected on them directly. There exist dozens of possible plots but only the most informative ones are sent. In Europe and the Mideast the position on the main scatter plot directly indicates geography. Otherwise, it is indirect. This is taken into account when the spot(s) on the geographic map plot are calculated. The test person (you) is marked with a white circle with either a crosshairs or a white dot at the center. The other colored dots are “comparison” people from various populations.
Mixed Ancestry and Related Details
Now if someone is of mixed ancestry, as said earlier, the spot(s) on the map, as well as PCA plots indicate the average. The quantitative lists (which sometimes are not sent if it is just one line like the common “English 100%”) represent possible combinations which average to the correct spot on all the PCA plots. Each line in the list is one set of possible combinations. We fit to 9 of the PCA plots, including ones sent and unsent. This is a “least squares fit”. A person, to fit, has to have the same weighted average position on all 9. Thus, say somebody lies in Romania on a Euro and Mideast only “main” one. They might actually be Romanian, or could be 60% Hungarian and 40% Jewish, or 75% French and 25% Iranian, or something more complicated. On other of the plots (not usually sent) however, Romania does not lie directly between Hungary and the Jews or between France and Iran, so the program (or an experienced person simply looking at a pile of plots) can tell what is correct. Well, most of the time it can but not always. For these cases, especially Jewish ancestry, I sometimes have separate special tests, whose results are reported in words.
People may wish to know “how far back in time do my tests go?” The answer is “forever”, in the sense that my tests all discuss single or mixed modern populations. For example modern populations such as North African ones can be mixtures several older ones (such as middle eastern and sub-Saharan African). This ancient mixture will appear on the chromosome plots. A similar situation exists with the Jews who are, on a long time scale, mixed Euro-Mideast. They will usually show spots of both Euro and Mideast on the chromosomes, even though we classify Jews as Mideastern.
If your parents are a typical English person and a typical Jew, that mix likely will appear on the list. But a mix of Irish, British, Romanian, and Palestinian could also appear, since these two possibilities share the same ancient roots. The spot on the map would be similar.
Small or modest amounts of, for example, Mideast or Eastern European in a basically English person can appear in the list to simply indicate that the person is from a point on the map east of England, such as Normandy, Germany, or Poland. Similar things happen on and across all continents, even Euro-East Asian. For example 97% European and 3% Han Chinese might simply be indicating a small eastward displacement on the map of the European component. If such a mix is truly recent, it almost always appears as such on the chromosomes. If ancient, it may or may not appear there. This fact is always very important to remember when interpreting the “quantitative” data listing.
Technical notes
The PCA plots use about 300,000 SNPs to create a (real symmetric) distance matrix which has dimension 200-1200 depending on how many populations are used. This is diagonalized, and the eigenvectors are what are plotted on the PCA graphs. The quantitative results are a fit to the first 9 "dimensions" of the full 1200x1200 matrix. The best few least squares fits to these
are reported as the “quantitative”, with the proviso that the least number of continental groups needed to "adequately" fit you is used. "Adequate" means that, averaged over the 9 dimensions, you fit within the area (actually 9-D volume) of a "typical" comparison population. The "continental" spots on the map are the average of the quantitative list, but more may be averaged than are sent in some cases. The green spot on the map is the average of the best 4-continental group fits and is usually (slightly) more accurate. This 4-population fit is seldom reported because it usually contains several populations with really tiny amounts.
The chromosome plots are an entirely separate calculation. They are a "maximum likelihood" estimate of the best two-continent fit over a sliding window using (for the autosomes) a Viterbi algorithm decision process. The "emission probabilities" of this are the probabilities of fitting 12-marker haplotypes. The block size for the "cut points" is 4 markers even though the average is over dozens of markers. The X is painted using fixed 12 marker blocks using a simple sliding window average, not Viterbi, and is less accurate. The computer program used is written by me and is not any of the “standard” ones such as “Structure”.
It should be emphasized that the accuracy of the results depends significantly on how representative the reference samples are in relation to the testee’s ancestry. We use some 90 groups of reference samples, all from academic studies. Not all available groups are used because for some of the analysis methods using too many groups can cause mathematical problems, and in a few cases the results appear simply silly when certain pairs are used.
Doug McDonald
I am a former professor at the University of Illinois at Urbana-Champaign. This work is neither supported nor endorsed by the University.
The list of populations we use as comparisons is
English
Irish
Russian
Basque
French
Italian
Sardinian
Tuscan
Spain
Romania
Chuvash
Belorus
Hungary
Lithuania
Finland
Adygei
Jewish
Mozabite
Bedouin
Bedouin South
Druze
Palestinian
Iranian
Egyptian
Moroccan
Turkish
Georgian
Cypriot
Armenian
Balochi
Brahui
Burusho
Hazara
Kalash
Makrani
Pathan
Sindhi
N_India
S_India
Uygur
Melanesian
Papuan
Columbian
Karitiana
Surui
Maya
Pima
Na-Dene
Eskimo
Bantu South Africa
Bantu Kenya
Mandenka
Yoruba
Biaka Pygmy
Mbuti Pygmy
San
Maasai
EthiopianOromo
EthiopianTigray
EthiopianAmhara
Han
Northern Han
Dai
Daur
Hezhen
Lahu
Miao
Oroqen
She
Tujia
Tu
Xibo
Yi
Mongolian
Naxi
Cambodian
Malaysian
Japanese
Yakut
Chukchi
Altai
Buryat
Dolgan
Evenk
Ket
Koryak
MongolianChinese
Nganassan
Selkup
Tuva

Timawa
08-24-2016, 04:50 PM
To understand it we'll have to know exactly how PCA plots work!..Dr McDonald also does PCA plot analysis. He mentions it in this guide:

Thanks for the guide. I'm trying to understand how it works.

Can you tell me which groups are Filipinos mostly closer to?

http://i45.photobucket.com/albums/f87/redwine88/srep26066-f5_zpssi01tr1r.jpg

sailormoon
08-25-2016, 02:30 AM
quite interesting that chinese have finnish


What the Finns and Chinese have in common is Y-DNA haplogroup N. 15.0% of Guangdong Han and 6.8% of Shaanxi Han belong to this haplogroup (Hammer 2005). Haplogroup N1c entered Europe about 12,00014,000 years ago from Asia and the initial prehistoric migration of haplogroup N began in southern China. Haplogroup N is extremely rare in Southeast Asian populations (<1.5%).



http://1.bp.blogspot.com/-npG1QhnACKE/UcPJm8EpJvI/AAAAAAAAI60/K_aLoq8Sptk/s1600/journal.pone.0066102.g004.png
The Y-chromosome haplogroup N-M231 (Hg N) is distributed widely in eastern and central Asia, Siberia, as well as in eastern and northern Europe. Previous studies suggested a counterclockwise prehistoric migration of Hg N from eastern Asia to eastern and northern Europe. However, the root of this Y chromosome lineage and its detailed dispersal pattern across eastern Asia are still unclear. We analyzed haplogroup profiles and phylogeographic patterns of 1,570 Hg N individuals from 20,826 males in 359 populations across Eurasia. We first genotyped 6,371 males from 169 populations in China and Cambodia, and generated data of 360 Hg N individuals, and then combined published data on 1,210 Hg N individuals from Japanese, Southeast Asian, Siberian, European and Central Asian populations. The results showed that the sub-haplogroups of Hg N have a distinct geographical distribution. The highest Y-STR diversity of the ancestral Hg N sub-haplogroups was observed in the southern part of mainland East Asia, and further phylogeographic analyses supports an origin of Hg N in southern China. Combined with previous data, we propose that the early northward dispersal of Hg N started from southern China about 21 thousand years ago (kya), expanding into northern China 1218 kya, and reaching further north to Siberia about 1214 kya before a population expansion and westward migration into Central Asia and eastern/northern Europe around 8.010.0 kya. This northward migration of Hg N likewise coincides with retreating ice sheets after the Last Glacial Maximum (2218 kya) in mainland East Asia.
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0066102

Ingenious
08-25-2016, 03:17 AM
Thanks for the guide. I'm trying to understand how it works.

Can you tell me which groups are Filipinos mostly closer to?

http://i45.photobucket.com/albums/f87/redwine88/srep26066-f5_zpssi01tr1r.jpg

How about we dont group up racially different people and question which group that hodgepodge is most closer to

Iloko
08-26-2016, 03:55 PM
Thanks for the guide. I'm trying to understand how it works.

Can you tell me which groups are Filipinos mostly closer to?

http://i45.photobucket.com/albums/f87/redwine88/srep26066-f5_zpssi01tr1r.jpg
I think Filipinos form their own cluster to be honest. We have some South Asian and Papuan admixture that pulls us a little closer to Eastern Indonesians than the Dais. Some Filipinos will also be a little closer to Han because of Northeast Asian admixture.

To me the plot looks at the overall average genetic distance of each individual sample like McDonald said, but I can't really say for sure; I could be wrong.

Timawa
09-06-2016, 09:19 PM
http://mbe.oxfordjournals.org/content/early/2012/07/28/molbev.mss169/F4.large.jpg

http://mbe.oxfordjournals.org/content/early/2012/07/28/molbev.mss169.full

Timawa
09-17-2016, 06:49 AM
Polynesians


Among self-reported East Asians, all had evidence of East Asian genetic ancestry; a sizeable proportion (21.7%) also had evidence of Pacific Islander genetic
ancestry, but this likely represents difficulty in differentiating East Asian and Pacific Islander genetic ancestry.

Source: http://www.genetics.org/content/early/2015/06/18/genetics.115.178616.full.pdf+html


In an analysis of the DNA of 1,000 individuals from 41 Pacific populations, an international team of scientists found strong evidence showing that Polynesians and Micronesians in the central and eastern islands had almost no genetic relationship to Melanesians, in the western islands like Papua New Guinea and the Bismarck and Solomons archipelagos.

The researchers also concluded that the genetic data showed that the Polynesians and Micronesians were most closely related to Taiwan Aborigines and East Asians. They said this supported the view that these migrating seafarers originated in Taiwan and coastal China at least 3,500 years ago.

Source: http://www.nytimes.com/2008/01/18/world/asia/18islands.html?_r=0


"Our study of the mtDNA evidence shows the interactions between the islands of Southeast Asia and the Pacific was far more complex than previous accounts tended to suggest and it paves the way for new theories of the spread of Austronesian languages."

Source: http://www.sciencedaily.com/releases/2011/02/110203124726.htm

Iloko
10-21-2016, 05:08 AM
A bunch of Filipino 23andme ancestry compositions:

http://i.imgur.com/gmhYjoE.jpg
...
http://i.imgur.com/2zrLmSZ.jpg
...
http://i.imgur.com/VWClSnQ.jpg
...
http://i.imgur.com/EMX34jI.jpg
...
http://i.imgur.com/a1NzdGf.jpg
...
http://i.imgur.com/TOSyf6N.jpg
...
http://i.imgur.com/SFLAVq7.jpg
...
http://i.imgur.com/oVhyF0Q.jpg
...
http://i.imgur.com/VdmYBp5.jpg
...
http://i.imgur.com/6wmXUnz.jpg
...
http://i.imgur.com/AeVMziK.jpg
...
http://i.imgur.com/MQVwkYL.jpg
...
http://i.imgur.com/xEG2NT9.jpg
...
http://i.imgur.com/GTMMobh.jpg
...
http://i.imgur.com/ymOGese.jpg
...
http://i.imgur.com/4RxXEGt.jpg
...
http://i.imgur.com/f5Wzo6I.jpg
...
http://i.imgur.com/ZUqQVy9.jpg
...
http://i.imgur.com/XCJwhIb.png

Iloko
10-21-2016, 05:08 AM
A bunch of Filipino 'puntDNAL K15' Gedmatch calculator results:

http://i.imgur.com/Youk7Ur.jpg
...
http://i.imgur.com/XBjMvPD.jpg
...
http://i.imgur.com/dCqslmi.jpg
...
http://i.imgur.com/6xgJZNh.jpg
...
http://i.imgur.com/E3lcGdZ.jpg
...
http://i.imgur.com/EGNO89Y.jpg
...
http://i.imgur.com/kxA5a93.jpg
...
http://i.imgur.com/3bKRSuZ.jpg
...
http://i.imgur.com/G3TTVZj.jpg
...
http://i.imgur.com/bFmVC6Z.jpg
...
http://i.imgur.com/68vpBRd.jpg
...
http://i.imgur.com/k695pJL.jpg
...
http://i.imgur.com/uNLv3WE.jpg

knowledge is king
03-27-2017, 08:34 PM
There was not a lot of discussion about Asians' genetic studies going here. I will post various charts and link them to articles.



What are the Chinese nations with highest C* and C3* distributions?

What is a Chinese province with highest C* percentage besides of Inner Mongolia?

Can someone clarify this?

knowledge is king
03-27-2017, 08:46 PM
Interesting. Genghis Khan and Y Haplogroup C.

https://pbs.twimg.com/media/CDmIUiOWEAAfy8m.png


https://pbs.twimg.com/media/B7DxXiyCQAA9JkG.png


https://pbs.twimg.com/media/C7unh9VVsAAqAyA.jpg

Timawa
03-29-2017, 06:53 AM
Interesting. Genghis Khan and Y Haplogroup C.

https://pbs.twimg.com/media/CDmIUiOWEAAfy8m.png


https://pbs.twimg.com/media/B7DxXiyCQAA9JkG.png


https://pbs.twimg.com/media/C7unh9VVsAAqAyA.jpg

Any speculations on why the haplogroup C is overrun by the newly emerged haplogroup O? The majority of EA/SEA's y-dna is O.

Timawa
03-29-2017, 07:03 AM
I've discovered something interesting now. There's a rare y-dna in my 23andme DNA relative list. I believe, the y-dna D.

knowledge is king
03-31-2017, 01:13 AM
A bunch of Filipino 'puntDNAL K15' Gedmatch calculator results:

http://i.imgur.com/Youk7Ur.jpg

Here is mine 'puntDNAL K15', would be nice to compare them.

My East Asian share is lower while Siberian one is higher. Also I have European component.

https://4.bp.blogspot.com/--bV4UthbQCo/WN2saVXafkI/AAAAAAAAGsQ/2jkDIJFNJlsDs_TDbosKhy4OqNmOjpDtwCLcB/s1600/Image-29.png

Pahli
03-31-2017, 01:21 AM
Here is mine 'puntDNAL K15', would be nice to compare them.

My East Asian share is lower while Siberian one is higher. Also I have European component.

https://4.bp.blogspot.com/--bV4UthbQCo/WN2saVXafkI/AAAAAAAAGsQ/2jkDIJFNJlsDs_TDbosKhy4OqNmOjpDtwCLcB/s1600/Image-29.png

Embrace your ancient Iranic and Mongolian roots :laugh:

Squall Leonhart
03-31-2017, 01:23 AM
Embrace your ancient Iranic and Mongolian roots :laugh:

I love you, Iranic daddy. :love0031:

knowledge is king
03-31-2017, 01:25 AM
Which share at this diagram should represent Iranic component in particular?

Squall Leonhart
03-31-2017, 01:27 AM
Which share at this diagram should represent Iranic component in particular?

I am misinformed but could be Tajiks/Pamiris?

Pahli
03-31-2017, 01:32 AM
Which share at this diagram should represent Iranic component in particular?

Iranic component would be Caucasian, NE_Euro and S_Indian, the Mongolian would be dominantly E_Asian.

But modern Iranic component is much more Caucasian now with the extinction of East Iranian nomads in Europe and the Steppes.

Iloko
03-31-2017, 01:34 AM
Here is mine 'puntDNAL K15', would be nice to compare them.

My East Asian share is lower while Siberian one is higher. Also I have European component.

https://4.bp.blogspot.com/--bV4UthbQCo/WN2saVXafkI/AAAAAAAAGsQ/2jkDIJFNJlsDs_TDbosKhy4OqNmOjpDtwCLcB/s1600/Image-29.png
Ahh so for your Mongoloid component it is not FULLY 100% Siberian since there is also that 37.28% 'E_Asian' as well. It will probably depend on the calculator though as it is relative to what reference population was selected for a certain component making that population the "pure" population.

On some GEDmatch calcs I score zero Siberian while on others 5-10%+

On this PCA Siberians seem distant enough to be their own race it looks, in between East Asians and Native Americans it appears:
http://2.bp.blogspot.com/-Va8sH8ho66k/Uo0OKBpX4fI/AAAAAAAAJZo/RLd9i4JYpv4/s1600/pca.png

knowledge is king
03-31-2017, 01:41 AM
I always wonder, what on the Graph B the axis values of PC1 and PC2 mean? What exact distribution we consider there consequently? Without proper understanding this picture have no sense.

knowledge is king
03-31-2017, 01:54 AM
Ahh so for your Mongoloid component it is not FULLY 100% Siberian since there is also that 37.28% 'E_Asian' as well. It will probably depend on the calculator though as it is relative to what reference population was selected for a certain component making that population the "pure" population.

On some GEDmatch calcs I score zero Siberian while on others 5-10%+

On this PCA Siberians seem distant enough to be their own race it looks, in between East Asians and Native Americans it appears:


I made a full database of my all admixture compositions I got from calculators on Gedmatch.

You can view my all admixture compositions for now right here :

zhīsh's all Eurogenes runs
https://docs.google.com/document/d/1cdmYqbjRFsbxrJK9Xe7axeHN_Buozg7lnHcE6MvmL5A/edit?usp=sharing

zhīsh's all GedrosiaDNA runs
https://docs.google.com/document/d/1vnosU8TxJP0Ug9gOnGIM9vPy0g2E6N_SgObhtinAj3w/edit?usp=sharing

zhīsh's all puntDNAL runs
https://docs.google.com/document/d/1DrwL_Zd1KGiPax-e77_p3jOFmU4nYSps0zYe7oAGYOo/edit?usp=sharing

zhīsh's all Dodecad runs
https://docs.google.com/document/d/1G-ggj7NzmKTyx7y6qkl8LTVOOOZCGdQGIC-by88OGTw/edit?usp=sharing

zhīsh's all MDLP runs
https://docs.google.com/document/d/17RDzcbWSASPqJuE247iskkIV34f_nVp_KS5v0fuAgYA/edit?usp=sharing

zhīsh's all Ethiohelix runs
https://docs.google.com/document/d/17orcaHpsgQIB7eidBdS7ebfefoc8k3QlM39YsUe_iHY/edit?usp=sharing

zhīsh's all HarappaWorld runs
https://docs.google.com/document/d/1pMiGAaiFuJP40aHNyuPWzFu9R4uM2U-eykY3rDnrny8/edit?usp=sharing

You are highly welcomed to see them and to express your opinion. :)

Pahli
03-31-2017, 02:04 AM
I made a full database of my all admixture compositions I got from calculators on Gedmatch.

You can view my all admixture compositions for now right here :

zhīsh's all Eurogenes runs
https://docs.google.com/document/d/1cdmYqbjRFsbxrJK9Xe7axeHN_Buozg7lnHcE6MvmL5A/edit?usp=sharing

zhīsh's all GedrosiaDNA runs
https://docs.google.com/document/d/1vnosU8TxJP0Ug9gOnGIM9vPy0g2E6N_SgObhtinAj3w/edit?usp=sharing

zhīsh's all puntDNAL runs
https://docs.google.com/document/d/1DrwL_Zd1KGiPax-e77_p3jOFmU4nYSps0zYe7oAGYOo/edit?usp=sharing

zhīsh's all Dodecad runs
https://docs.google.com/document/d/1G-ggj7NzmKTyx7y6qkl8LTVOOOZCGdQGIC-by88OGTw/edit?usp=sharing

zhīsh's all MDLP runs
https://docs.google.com/document/d/17RDzcbWSASPqJuE247iskkIV34f_nVp_KS5v0fuAgYA/edit?usp=sharing

zhīsh's all Ethiohelix runs
https://docs.google.com/document/d/17orcaHpsgQIB7eidBdS7ebfefoc8k3QlM39YsUe_iHY/edit?usp=sharing

zhīsh's all HarappaWorld runs
https://docs.google.com/document/d/1pMiGAaiFuJP40aHNyuPWzFu9R4uM2U-eykY3rDnrny8/edit?usp=sharing

You are highly welcomed to see them and to express your opinion. :)

Should compare to this Gkturk sample from around 600 A.D:

# Population Percent
1 Northeast_Asian 34.66
2 Southeast_Asian 27.53
3 West_European 14.43
4 East_European 10.81
5 West_Asian 10.54
6 South_Asian 1.82


Finished reading population data. 227 populations found.
12 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Altai @ 9.495087
2 Kyrgyzstani @ 11.248915
3 Mongol @ 18.637209
4 Tuva @ 19.664129
5 Uygur @ 20.699966
6 Hazara @ 21.404951
7 Dolgan @ 22.143494
8 Buryat @ 24.513571
9 Uzbeks @ 24.735023
10 Buryat @ 24.821211
11 Yukagir @ 25.699213
12 Selkup @ 29.646893
13 Yakut @ 31.198229
14 Oroqen @ 32.852871
15 Evenk_15 @ 36.918957
16 Daur @ 37.456837
17 Mongola @ 37.687393
18 Hezhen @ 37.980820
19 Xibo @ 41.094051
20 Chuvashs_16 @ 43.942310


Gokturk sample, MDLP World 22

# Population Percent
1 East-Siberean 27.40
2 East-South-Asian 17.19
3 North-East-European 14.37
4 Samoedic 13.63
5 North-Siberean 10.45
6 West-Asian 8.94
7 South-America_Amerind 1.88
8 Indo-Iranian 1.25
9 Paleo-Siberian 1.16


Finished reading population data. 276 populations found.
22 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Altaic @ 8.787984
2 Hakas @ 9.468202
3 Kazakh @ 11.127083
4 Shor @ 13.193381
5 Karakalpak @ 13.344393
6 Kyrgyz @ 14.426857
7 Uygur @ 20.866812
8 Tuva @ 21.522612
9 Bashkir @ 22.600176
10 Kalmyk @ 22.675993
11 Hazara @ 22.718258
12 Mongol @ 25.225319
13 Uzbek @ 25.446110
14 Buryat @ 25.842859
15 Nivhi @ 28.907421
16 Ste7 @ 29.818701
17 Yukagir @ 30.685778
18 Yukagir @ 30.685778
19 Oroqen @ 33.707439
20 Tatar_Lithuania @ 33.784519

Using 2 populations approximation:
1 50% Bashkir +50% Tuva @ 6.672640

Squall Leonhart
03-31-2017, 02:08 AM
Should compare to this Gkturk sample from around 600 A.D:

# Population Percent
1 Northeast_Asian 34.66
2 Southeast_Asian 27.53
3 West_European 14.43
4 East_European 10.81
5 West_Asian 10.54
6 South_Asian 1.82


Finished reading population data. 227 populations found.
12 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Altai @ 9.495087
2 Kyrgyzstani @ 11.248915
3 Mongol @ 18.637209
4 Tuva @ 19.664129
5 Uygur @ 20.699966
6 Hazara @ 21.404951
7 Dolgan @ 22.143494
8 Buryat @ 24.513571
9 Uzbeks @ 24.735023
10 Buryat @ 24.821211
11 Yukagir @ 25.699213
12 Selkup @ 29.646893
13 Yakut @ 31.198229
14 Oroqen @ 32.852871
15 Evenk_15 @ 36.918957
16 Daur @ 37.456837
17 Mongola @ 37.687393
18 Hezhen @ 37.980820
19 Xibo @ 41.094051
20 Chuvashs_16 @ 43.942310


Gokturk sample, MDLP World 22

# Population Percent
1 East-Siberean 27.40
2 East-South-Asian 17.19
3 North-East-European 14.37
4 Samoedic 13.63
5 North-Siberean 10.45
6 West-Asian 8.94
7 South-America_Amerind 1.88
8 Indo-Iranian 1.25
9 Paleo-Siberian 1.16


Finished reading population data. 276 populations found.
22 components mode.

--------------------------------

Least-squares method.

Using 1 population approximation:
1 Altaic @ 8.787984
2 Hakas @ 9.468202
3 Kazakh @ 11.127083
4 Shor @ 13.193381
5 Karakalpak @ 13.344393
6 Kyrgyz @ 14.426857
7 Uygur @ 20.866812
8 Tuva @ 21.522612
9 Bashkir @ 22.600176
10 Kalmyk @ 22.675993
11 Hazara @ 22.718258
12 Mongol @ 25.225319
13 Uzbek @ 25.446110
14 Buryat @ 25.842859
15 Nivhi @ 28.907421
16 Ste7 @ 29.818701
17 Yukagir @ 30.685778
18 Yukagir @ 30.685778
19 Oroqen @ 33.707439
20 Tatar_Lithuania @ 33.784519

Using 2 populations approximation:
1 50% Bashkir +50% Tuva @ 6.672640

+1

Sikeliot
03-31-2017, 02:29 AM
On this PCA Siberians seem distant enough to be their own race it looks, in between East Asians and Native Americans it appears:
http://2.bp.blogspot.com/-Va8sH8ho66k/Uo0OKBpX4fI/AAAAAAAAJZo/RLd9i4JYpv4/s1600/pca.png

Who are the South Asians near the Melanesians? Are those probably South Indians, Sri Lankans and Bengalis?

Squall Leonhart
03-31-2017, 02:40 AM
Who are the South Asians near the Melanesians? Are those probably South Indians, Sri Lankans and Bengalis?

I guess so.

Maguzanci
03-31-2017, 06:02 PM
Ahh so for your Mongoloid component it is not FULLY 100% Siberian since there is also that 37.28% 'E_Asian' as well. It will probably depend on the calculator though as it is relative to what reference population was selected for a certain component making that population the "pure" population.

On some GEDmatch calcs I score zero Siberian while on others 5-10%+

On this PCA Siberians seem distant enough to be their own race it looks, in between East Asians and Native Americans it appears:
http://2.bp.blogspot.com/-Va8sH8ho66k/Uo0OKBpX4fI/AAAAAAAAJZo/RLd9i4JYpv4/s1600/pca.png

Based on PCA, it seems like East Asian are as distant from Native Americans as they are from Europeans, Central and South Asians. This makes a lot of sense when looking it through FST datas.

Timawa
04-24-2017, 05:52 PM
https://bloximages.newyork1.vip.townnews.com/postguam.com/content/tncms/assets/v3/editorial/2/28/228b1204-27fe-11e7-8743-5bd0417ca239/58fc6499269e9.image.jpg

His y-dna and mtdna
https://bloximages.newyork1.vip.townnews.com/postguam.com/content/tncms/assets/v3/editorial/0/9a/09a616d0-27fe-11e7-b77b-dff1f41b5b95/58fc64778a948.image.jpg

Chamorro Roots researcher Bernard Punzalan uses historical records to develop the massive archive that helps thousands of Guam's native inhabitants discover their ancestry. Recently DNA sampling has added another layer to Punzalan's family history, and the roots of his tree, he discovered, go much, much deeper.

Chamorro Roots is a digital archive of more than 350,000 records compiled by Punzalan in recent years. It started as a means of self-discovery before becoming a widely used resource for Chamorros at home and abroad.

"I started out just doing the research for myself and my sons. I wanted to leave something behind for them," Punzalan said. "But then as I was digging into records from before the 1800s things started getting difficult. I realized that If I was having this problem, there are probably others that are hitting this wall also."

In 2015, Miguel Vilar, project manager and scientific lead for National Geographic Society's Genographic Project, delivered a lecture at the University of Guam about the known origins of the indigenous people of the Marianas as indicated by connections made through samples of Chamorros' DNA.

'Threw caution to the wind'

Punzalan had long been opposed to submitting his DNA for testing, citing identity theft concerns. But in December last year, after being enticed by a discounted Ancenstry.com package, the U.S. Army veteran finally buckled.

"The price went down from $99 to $69 on Ancestry.com and I sort just threw caution to the wind," Punzalan said. "But what's great about it is you can extract the raw data and plug it into other genealogy systems. You can't do that with National Geographic's data, which I've also used."

Like many participants in these genealogy tests, Punzalan was surprised to learn that his DNA shared unexpected similarities with people from a number of different regions.

For instance, one screenshot provided by Punzalan shows that his DNA shares less-than-1 percent similarities with peoples from Scandinavia and Northwest Russia.

27-percent Polynesian

But he was most amazed by another set of data.

"I was actually more surprised with my ancestry's 27 percent Pacific Islander finding, and the data pegging me as Polynesian," Punzalan said. "I think what this demonstrates is that more people from the Marianas need to be tested to power up the numbers and geography."

According to the Genographic Project, maternal haplogroups E1 and E2 are found most frequently in the Marianas. Punzalan belongs to branch E2A, which emerged from Sundaland, present-day Indonesia, some 15,000 years ago.

While DNA genealogy data is interesting in its exploration of human migration, Punzalan said it should be utilized with careful scrutiny.

Many Pacific Islander scholars are against "blood quantum," the fractionalizing of one's "blood amount," designations of identity.

In Hawaii for instance, the legal definition of "native Hawaiian" is a "descendant with at least one-half blood quantum of individuals inhabiting the Hawaiian Islands prior to 1778," according to the Hawaiian Homes Commission Act of 1920.

But the problem with this logic is that it supposes that there was ever a "full-blooded Hawaiian," and that if a person is less-than one-half Hawaiian, they will no longer be eligible for distinct Native rights, chiefly, in this case, designated Hawaiian lands.

"I used to think blood quantum was important," Punzalan said. "The Punzalan name is Filipino, and for many years I struggled to come to terms with that and try to 'prove' how Chamorro I was."

"But I realized that's it's not about blood. There's a spiritual energy Chamorros have, as well as other islanders, that makes us who we are. Blood and DNA are Western concepts of race. We shouldn't let those things determine who we are," Punzalan said.

In the meantime, Punzalan said he thinks DNA will help "fill in the gaps" in ancestral data pre-1898.

Link: https://www.postguam.com/news/local/roots-researcher-discovers-dna-origins/article_f2ce4d98-27fc-11e7-959b-e3d47872e59f.html

Timawa
04-24-2017, 05:54 PM
The lead scientist of a study examining the genetics of Chamorros said it was “a huge surprise” when researchers discovered a connection between Native Americans and the indigenous people of Guam and nearby Pacific islands.

Miguel Vilar, a molecular anthropologist and science writer, presented his findings in a lecture Tuesday at the University of Guam.

Vilar is the science manager for National Geographic’s Genographic Project, an initiative that seeks to use DNA from participants to discover human genetic roots.

No seat was left vacant in the crowded University lecture hall during his speech. Remaining audience members had to stand elbow-to-elbow along the sides of the room.

“At first I couldn’t figure out what it was,” Vilar said, describing the discovery of Native American ancestry in Chamorro DNA. “I was thinking they were actually Native Americans that migrated over here.”
Dr. Miguel Vilar, project manager and scientific lead

(Photo: Mark Scott/PDN)

But history books pointed to a more likely answer, he added. Long ago, people from the Mexico region came to Guam, and depending on when they came, they could’ve already been part of a mixed European-Native American gene pool, he said.

“Those were the people coming over and that would explain the Native American (genes),” he said. “But it was unexpected.”

The study examined DNA from 200 participants that were collected in two separate time periods — the first being in the ’90s and ’00s. In that timeframe, DNA was collected from 122 individuals. Eighty-five were from Guam, 31 were from Saipan and six were from Rota.

The participants from the sample were Chamorro, Carolinian and a combination of the two.

In 2013, about 85 new participants from Guam submitted DNA samples. About 69 of the new samples were Chamorros.

Results from the study confirmed linguistic and archaeological evidence that Chamorros originated somewhere in Eastern Indonesia, Vilar said.

And, additionally, Vilar said the data showed Native American ancestry in Chamorro DNA.

That finding was “a huge surprise,” he said.
Documentary about Chamorros released on DVD, Blu-ray

In just about all of the strains of Chamorro DNA, Vilar said findings showed 3 to 4 percent of Native American ancestry.

“Some were as high as 7, some, 0 to 1 percent,” he said. “Three to 4 percent was common.”

Vilar said results of his research found Mexicans had about 50 to 70 percent Native American ancestry.

“Spanish people were bringing people over from Mexico by the 17th century, mixing with the Native American population in Mexico,” he said.

Research findings also concluded that, on average, each Chamorro person in the study had about 20 percent European ancestry.

Vilar said it’s unclear what the DNA of a person who’s 100 percent Chamorro would look like at this point because of scientific and technological limitations.

Added insight into Chamorro DNA could be found if more Chamorros participated in the project, he said.

To find out more about the project and how to participate, individuals can visit http:///203X9HU.

Vilar was the 29th speaker in the UOG presidential series lecture. The series has featured distinguished guest speakers who have shared their knowledge on various topics with the university community.

The lecture will be available on the University of Guam website next week, said university spokesman Jonas Macapinlac.

Link: http://www.guampdn.com/story/news/local/2016/01/27/researcher-chamorros-have-native-american-ancestors/79390316/

DaiViet
12-02-2017, 03:53 PM
I think the terms NE Asian, East Asian, and SE Asian are more useful for geographic identification than for genetic grouping.

The thing that’s apparent is there are genetic clusters between Asians. NE Asian ethnicities can actually be genetically far from each other. The same is true with SE Asians. For instance, the genetic distance between Beijing Han to Japanese is even further than the distance between Beijing Han and Vietnamese — quite far. Likewise the distance between Vietnamese and Cambodian is equivalent to Vietnamese and Beijing Han, also far. Most Filipinos are very far from mainland SE Asians and even further from NE Asians. Malay and Indonesians as well.

Han Chinese form a long continuum, as they’re not as tightly clustered as most other ethnicities in continental Asia.

The table below compares genetic distance between ethnicities to the left and top. The larger the number the greater the distance.

http://i65.tinypic.com/t6pij8.png

In order of closeness for Beijing Chinese:
1. Southern Chinese
2. Vietnamese
3. Japanese
4. Thai
5. Cambodian
6. Filipino
7. Indonesian
8. Iban
9. Malaysian
10. Taiwanese aborigine

In order of closeness for Vietnamese:
1. Southern Chinese
2. Thai
3. Beijing Chinese, Cambodian
4. Filipino
5. Indonesian
6. Japan, Malaysian, Iban
7. Taiwanese aborigine

Southern Chinese straddle between Beijing Chinese and Vietnamese, being equidistant between both. The southern Han Chinese samples typically used are from 1000 Genomes where they were taken from Fujian and Hunan province.

KuakoVoice
08-10-2020, 01:50 AM
There was not a lot of discussion about Asians' genetic studies going here. I will post various charts and link them to articles.

http://i45.photobucket.com/albums/f87/redwine88/Mobile%20Uploads/5ADDE87E-F10F-49B0-867B-551607BCFF5A_zps8ltjsjzo.jpg

http://www.nature.com/articles/srep14375

http://i45.photobucket.com/albums/f87/redwine88/srep26066-f5_zpssi01tr1r.jpg

http://www.nature.com/articles/srep26066#f5

Since this graph totally ignored Eastern Indonesia and Timor Leste (again) here are some more genetic studies to finish it.

https://ibb.co/VMmqY6q
https://ibb.co/jkcHF2K