Dilawer's SAPDA vs Davidski's calculators

[Zoro]

Davidski whoever he is at Eurogenes seems to have an ego problem and always talks shit about other bloggers and researchers like Dilawer of Eurasiandna.com. I was reading an article at https://eurasiandna.com/sapda-admixture-inference-updates-revisions/ about the SAPDA test and noticed that there was a comparison chart showing SAPDA results vs Eurogenes calculator results for Neanderthal.

I noticed that Eurogenes K13 showed Altai Neanderthal as something like 96% related to Africans whereas Dilawer's SAPDA test showed it only about 25% related to Africans and about 75% related to Eurasians and mostly to East Eurasians.

Based on what I have read everywhere Neanderthal left Africa a few hundred thousand years ago but admixed with Eurasians more recently in Europe and Asia. So to me Davidski's calculator doesn't make any sense and he should not be talking shit about Dilawer's test.

What do you guys think does Davidski got it right and Altai Neanderthal is 97% related to Africans and only 3% to Eurasians or do you think Dilawer's SAPDA makes more sense showing Altai Neanderthal about 75% related to Eurasians (mostly East Eurasians) ?

I don't have the G25 results for Altai Neanderthal but if someone does feel free to post it.

https://i.imgur.com/o62c0kO.jpg

[bulunulu]

Not sure if this is a good idea to post such thing, dude

seems like to start riling up a fight between followers of DIlawer(much fewer) and Davidski(way more people)

[JamesBond007]

Davidski whoever he is at Eurogenes seems to have an ego problem and always talks shit about other bloggers and researchers like Dilawer of Eurasiandna.com. I was reading an article at https://eurasiandna.com/sapda-admixture-inference-updates-revisions/ about the SAPDA test and noticed that there was a comparison chart showing SAPDA results vs Eurogenes calculator results for Neanderthal.

I noticed that Eurogenes K13 showed Altai Neanderthal as something like 96% related to Africans whereas Dilawer's SAPDA test showed it only about 25% related to Africans and about 75% related to Eurasians and mostly to East Eurasians.

Based on what I have read everywhere Neanderthal left Africa a few hundred thousand years ago but admixed with Eurasians more recently in Europe and Asia. So to me Davidski's calculator doesn't make any sense and he should not be talking shit about Dilawer's test.

What do you guys think does Davidski got it right and Altai Neanderthal is 97% related to Africans and only 3% to Eurasians or do you think Dilawer's SAPDA makes more sense showing Altai Neanderthal about 75% related to Eurasians (mostly East Eurasians) ?

I don't have the G25 results for Altai Neanderthal but if someone does feel free to post it.



Davidksi does not know what he is talking about. Neanderthals in general mixed with Europeans and middle Easterners or West Eurasians, Homo Erectus mixed with Asians and if any earlier species of homo mixed with Africans it was Homo Rudolfensis.

[Zoro]

Not sure if this is a good idea to post such thing, dude

seems like to start riling up a fight between followers of DIlawer(much fewer) and Davidski(way more people)

Dude I'm not asking who is better looking or wiser. It's a valid question as to whether Davidski's results for the Altai Neanderthal make more sense or Dilawer's.

Now if he goes around shitting on others work he deserves to have the truth pointed out. I remember he did it with Dienekes, the creator of Dodecad. As they say what goes around comes around haha.

BTW Dilawer has plenty of followers. His followers are more of the kind that really know how this stuff works. Quality vs quantity if you want to call it that. Davidski is a good blogger but Dilawer is the one that writes his own software and digs really deep into the scientific stuff IMO.

[bulunulu]

Dude I'm not asking who is better looking or wiser. It's a valid question as to whether Davidski's results for the Altai Neanderthal make more sense or Dilawer's.

Now if he goes around shitting on others work he deserves to have the truth pointed out. As they say what goes around comes around haha.

BTW Dilawer has plenty of followers. His followers are more of the kind that really know how this stuff works. Quality vs quantity if you want to call it that. Davidski is a good blogger but Dilawer is the one that writes his own software and digs really deep into the scientific stuff IMO.

Dilawer really have many followers?

THe only ones i see are on yourdnaportal forum and thats just like 2-3 people + you(and some other iraqi girl maybe)

[Dick]

Davidksi does not know what he is talking about. Neanderthals in general mixed with Europeans and middle Easterners or West Eurasians, Homo Erectus mixed with Asians and if any earlier species of homo mixed with Africans it was Homo Rudolfensis.

This

[Adamastor]

Calculators look at different things based on different perspectives, these calcs were not made to pre-human populations anyway. Actually most pre-Homo Sapiens DNA would be closer to Mbuti Pygmies because they are the closest living people to Human Root. It doesn't matter with whom they mixed after.

[Zoro]

Dilawer really have many followers?

THe only ones i see are on yourdnaportal forum and thats just like 2-3 people + you(and some other iraqi girl maybe)

Dude you have no idea what you are talking about. His calculators and tests on Geneplaza are pretty popular. He also mentioned once that he gets thousands of visitors on his website from all over the world. His only problem I think is that the stuff he writes is much more difficult to understand than Davidski's blogging by ordinary people. So his followers are different from your average forum people.

[bulunulu]

Dude you have no idea what you are talking about. His calculators and tests on Geneplaza are pretty popular. He also mentioned once that he gets thousands of visitors on his website from all over the world. His only problem I think is that the stuff he writes is much more difficult to understand than Davidski's blogging by ordinary people. So his followers are different from your average forum people.

WHich website? Cant be eurasiandna?

[Zoro]

Calculators look at different things based on different perspectives, these calcs were not made to pre-human populations anyway. Actually most pre-Homo Sapiens DNA would be closer to Mbuti Pygmies because they are the closest living people to Human Root. It doesn't matter with whom they mixed after.

So you're saying that Altai Neanderthal shares more alleles with Africans than Eurasians. Post just one paper that shares your views. Good luck buddy.

[Zoro]

Dick and Jamesbond007, this is a poll question. I appreciate it if you check off the vote that shows your position.

[Cumansky]

Neither

[Zoro]

Neither

Ok but which result do you think makes more sense if you had to pick from the 2. Dilawer's showing Altai Neanderthal is much more related to Eurasians or Davidski's result showing it is much more related to Africans.

[Cumansky]

They are just systems, nothing is perfect, this is DNA game

Both geneticist produce parallel work IMO, both are trustworthy geneticist compared to mainstream companies

[J. Ketch]

Dude I'm not asking who is better looking or wiser. It's a valid question as to whether Davidski's results for the Altai Neanderthal make more sense or Dilawer's.

Now if he goes around shitting on others work he deserves to have the truth pointed out. I remember he did it with Dienekes, the creator of Dodecad. As they say what goes around comes around haha.

BTW Dilawer has plenty of followers. His followers are more of the kind that really know how this stuff works. Quality vs quantity if you want to call it that. Davidski is a good blogger but Dilawer is the one that writes his own software and digs really deep into the scientific stuff IMO.
You wouldn't happen to be the same ethnicity as the Dilawer character?

Is he responsible for the ridiculous Ancient Origins feature on FTDNA?
https://i.postimg.cc/Mp3Myz8H/Opera-Snapshot-2019-12-24-105250-www-familytreedna-com.png

[Dick]

You wouldn't happen to be the same ethnicity as the Dilawer character?

Is he responsible for the ridiculous Ancient Origins feature on FTDNA?
https://i.postimg.cc/Mp3Myz8H/Opera-Snapshot-2019-12-24-105250-www-familytreedna-com.png

https://www.theapricity.com/forum/showthread.php?207074-Indians-!-are-Yamnaya-(FTDNA)

[Imperator Biff]

You wouldn't happen to be the same ethnicity as the Dilawer character?

Is he responsible for the ridiculous Ancient Origins feature on FTDNA?
https://i.postimg.cc/Mp3Myz8H/Opera-Snapshot-2019-12-24-105250-www-familytreedna-com.png
That was done by razib khan of the gnxp blog.

[J. Ketch]

That was done by razib khan of the gnxp blog.
Ok, wrong Khan.

Pretty sure this was his calculator though
https://i.postimg.cc/rFDWpd4s/geneplaza.png

[Imperator Biff]

You wouldn't happen to be the same ethnicity as the Dilawer character?

Is he responsible for the ridiculous Ancient Origins feature on FTDNA?
https://i.postimg.cc/Mp3Myz8H/Opera-Snapshot-2019-12-24-105250-www-familytreedna-com.png
Double post*

[Imperator Biff]

Ok, wrong Khan.

Pretty sure this was his calculator though
https://i.postimg.cc/rFDWpd4s/geneplaza.png

Yeah it was, no idea why he lumped all the beakers together.....

[Zoro]

You wouldn't happen to be the same ethnicity as the Dilawer character?

Is he responsible for the ridiculous Ancient Origins feature on FTDNA?
https://i.postimg.cc/Mp3Myz8H/Opera-Snapshot-2019-12-24-105250-www-familytreedna-com.png

What exactly is his ethnicity and how is this relevant to the poll question namely does A or B make more sense

[J. Ketch]

What exactly is his ethnicity and how is this relevant to the poll question namely does A or B make more sense
Well he went by 'Kurd' on Anthrogenica, so I assume that refers to his ethnicity. He also started threads with large and bolded text like you have.

https://anthrogenica.com/showthread.php?13670-Upcoming-K15-Ancient-Cultures-Calculator-at-GenePlaza
https://anthrogenica.com/showthread.php?12934-K25-Updated-to-K29-Admixture-Calculator

[Dick]

Well he went by 'Kurd' on Anthrogenica, so I assume that refers to his ethnicity. He also started threads with large and bolded text like you have.

https://anthrogenica.com/showthread.php?13670-Upcoming-K15-Ancient-Cultures-Calculator-at-GenePlaza
https://anthrogenica.com/showthread.php?12934-K25-Updated-to-K29-Admixture-Calculator

Isnt Kurd the author of Gedrosia calculators on gedmatch. Davidski mentioned in his blog somewhere on the past that there's no such component as gedrosia

[Adamastor]

Isnt Kurd the author of Gedrosia calculators on gedmatch. Davidski mentioned in his blog somewhere on the past that there's no such component as gedrosia

Most of the calculators developed by MENAs and South Asians are a complete waste of time. Just stick to Eurogenes and G25 and you would be fine.

[Dick]

Most of the calculators developed by MENAs and South Asians are a complete waste of time. Just stick to Eurogenes and G25 and you would be fine.

I Dont know. PuntDnal is a good one for example

[Zoro]

Well he went by 'Kurd' on Anthrogenica, so I assume that refers to his ethnicity. He also started threads with large and bolded text like you have.

https://anthrogenica.com/showthread.php?13670-Upcoming-K15-Ancient-Cultures-Calculator-at-GenePlaza
https://anthrogenica.com/showthread.php?12934-K25-Updated-to-K29-Admixture-Calculator

Lol kurd is the name of a brahui tribe man and lots of baloch have last name kurd. They are the kurds that migrated from kurdistan into balochistan 500 years ago. Check out online

[Zoro]

Isnt Kurd the author of Gedrosia calculators on gedmatch. Davidski mentioned in his blog somewhere on the past that there's no such component as gedrosia

Gedrosia is the ancient name for balochistan area https://en.wikipedia.org/wiki/Gedrosia

[Zoro]

Most of the calculators developed by MENAs and South Asians are a complete waste of time. Just stick to Eurogenes and G25 and you would be fine.

Typical eurocentric view. So you’re saying that Eurogenes k13 showing Altai Neanderthal 97% African is good but SAPDA showing the sample as 75% Eurasian is a waste of time lol

[Zoro]

Davidski mentioned in his blog somewhere on the past that there's no such component as gedrosia

He thinks we’re all stupid and he can get away with his double standards or nonsense. Then why does he have SC Asian component in his calculator based on baluch. If we were to believe that there’s no gedrosian component based on baluch and brahui, then we also should believe that there’s no caucasian component based on georgians or no Amerindian component based on karitiana and so on

[vbnetkhio]

I noticed that Eurogenes K13 showed Altai Neanderthal as something like 96% related to Africans whereas Dilawer's SAPDA test showed it only about 25% related to Africans and about 75% related to Eurasians and mostly to East Eurasians.


it's not SAPDA vs. Davidski, it's SAPDA vs. Admixture, which wasn't made by Davidski. He just made a dozen of calculators in Admixture.

And Admixture is better imo. it's clear what it shows: ancestry, not relatedness. What components should the Neanderthal score in your opinion? Baltic and Atlantic? those components are both a specific mix of WHG, EHG, CHG and EEF, does the Neanderthal have ancestry from any of those groups? No. They had ancestry from earlier Neanderthals, and ultimatly from Homo Heidelbergensis. But there is no Neanderthal or H. Heidelbergensis component in the calcualtors, so the algorithm picks the most similar available component instead, the Pygmies.

We shouldn't be discussing is Admixture good, but why are Pygmies genetically similar to primitive Homo species? I would like to hear a good scientific explanation of that.

[Lucas]

it's not SAPDA vs. Davidski, it's SAPDA vs. Admixture, which wasn't made by Davidski. He just made a dozen of calculators in Admixture.

And Admixture is better imo. it's clear what it shows: ancestry, not relatedness. What components should the Neanderthal score in your opinion? Baltic and Atlantic? those components are both a specific mix of WHG, EHG, CHG and EEF, does the Neanderthal have ancestry from any of those groups? No. They had ancestry from earlier Neanderthals, and ultimatly from Homo Heidelbergensis. But there is no Neanderthal or H. Heidelbergensis component in the calcualtors, so the algorithm picks the most similar available component instead, the Pygmies.

We shouldn't be discussing is Admixture good, but why are Pygmies genetically similar to primitive Homo species? I would like to hear a good scientific explanation of that.

Once I uploaded to Gedmatch gorilla, chimp. macaque and marmosete genomes. Guess what components they scored?

[Zoro]

it's not SAPDA vs. Davidski, it's SAPDA vs. Admixture, which wasn't made by Davidski. He just made a dozen of calculators in Admixture.

And Admixture is better imo. it's clear what it shows: ancestry, not relatedness. What components should the Neanderthal score in your opinion?


Certainly not African. Altai Neanderthal should score Eurasian and specifically more E. Eurasian just like SAPDA shows. So in Eurogenes k13 it should score mostly E.Asian (Han, mongolian, siberian.....) since Altai admixed more with E. Asians and also some W. Asian whether it is caucasian W. Asian WHG .....but certainly not African.




We shouldn't be discussing is Admixture good, but why are Pygmies genetically similar to primitive Homo species? I would like to hear a good scientific explanation of that.

They are NOT. This is what you are not understanding. You are saying this based on ADMIXTURE calculator results but the TRUTH is they are not. If you were to do a whole genome sequence on Neanderthal, Africans, Eurasians guess what you’ll find ? You will shockingly find that Neanderthal shares more alleles with Eurasians than Africans!! You can dig up some papers that did whole genome sequence and see for yourself. So SAPDA is correct. It’s called ascertainment bias. The few hundred thousand SNPs DNA companies use are biased to show variation in mostly Europeans not Africans or S. Or E. Asians.

That’s the problem with most people. They don’t understand Dilawer because his stuff is too technical . Here’s an excerpt from Dilawer. Please read it slowly and careful so you can digest it


Eurasians, whether Europeans or Asians, are genetically very similar to each other, with genetic similarities far outweighing any differences they have. Most Eurasians have descended from a few thousand individuals who roamed Eurasia during the Upper Paleolithic and thus share a huge amount of DNA. Thus an admixture inference program needs to determine which mutations are shared due to distant common origins and which ones due to more recent introgression.

It is for this reason that some bioinformatic software such as qpDstat, qp3pop, and qpAdm, which are included in the ADMIXTOOLS software suite available at Reich Lab, use outgroups to filter out older common ancestral alleles.

An extreme example of how ancestral alleles and SNP ascertainment bias distort admixture calculations using the program ADMIXTURE is Neanderthal. When the Neanderthal is processed with an ADMIXTURE based calculator, regardless of calculator design, the results show it to be around 90% African. Of course we know this inference is incorrect because the Neanderthal lineage diverged from humans over 500,000 years ago, with some hybridization with humans occurring around 50,000 years ago in Eurasia and NOT in Africa.

To illustrate this problem, a comparison is made of ADMIXTURE program based calculator results from Gedmatch.com with SAPDA calculator results for a higher coverage Altai Neanderthal genome with Gedmatch ID XV3025795.

It is expected that a calculator output show that Neanderthal is equally related to Eurasians and Africans, with a slight shift towards Eurasians, due to its limited admixture with Eurasians around 50,000 years ago, long after its split from the ancestral Neanderthal-Human lineage. However, we see this not to be the case.

Figure 1 shows that except for the SAPDA program, the ADMIXTURE program based various calculator projects erroneously indicate that the Neanderthal sample is around 90% African.

By contrast, SAPDA indicates greater shared drift between Eurasians and the Altai Neanderthal sample vs Altai Neanderthal and Africans, specifically, Siberians, American Indians, and East Asian Eurasians. This of course is a more expected result than the results using ADMIXTURE software, which tend to show the Altai Neanderthal sample as approximately 90% African.

[Zoro]

Once I uploaded to Gedmatch gorilla, chimp. macaque and marmosete genomes. Guess what components they scored?


Let me guess. African!

[vbnetkhio]

Certainly not African. Altai Neanderthal should score Eurasian and specifically more E. Eurasian just like SAPDA shows. So in Eurogenes k13 it should score mostly E.Asian (Han, mongolian, siberian.....) since Altai admixed more with E. Asians and also some W. Asian whether it is caucasian W. Asian WHG .....but certainly not African.

i read that already. what Dilawer didn't mention is that admixture doesn't look at small pieces of inherited dna. they don't mean anything here. it's a modelling software, kind of like nmonte, but works directly with genes and not with dna results.

let's put it like this: you want to model the ancestry of a Mongolian. you give the algorithm 2 populations to draw admixtures from:
1)a group of Russians with 0.5% actual, real Mongolian ancestry.
2)a group of Chinese, with no Mongolian ancestry, zero, none whatsoever

what do you expect the Mongolian to score?

[Voskos]

SAPDA made sense for me, both the admixture scores and the GSI.

[Zoro]

i read that already. what Dilawer didn't mention is that admixture doesn't look at small pieces of inherited dna. they don't mean anything here. it's a modelling software, kind of like nmonte, but works directly with genes and not with dna results.

let's put it like this: you want to model the ancestry of a Mongolian. you give the algorithm 2 populations to draw admixtures from:
1)a group of Russians with 0.5% actual, real Mongolian ancestry.
2)a group of Chinese, with no Mongolian ancestry, zero, none whatsoever

what do you expect the Mongolian to score?

I don’t what you are saying. Admixture looks at individual SNPs not haplotypes like 23andme. So Admixture will pick up very ancient ancestry as well. Also Admixture can’t tell you if the score is because of ancient shared ancestry or because of recent direct geneflow. That’s why Europeans will score alot of yamnaya becuse yamnaya and Europeans have alot of EHG.

Admixture can’t figure out your scenario. Why don’t you run the Mongolian and Russian through one of Eurogenes calculator and see who scores more chinese. It sure won’t be the Russian you mention.Feel free to share their results

[vbnetkhio]

I don’t what you are saying. Admixture looks at individual SNPs not haplotypes like 23andme.
23andme, ancestry and everybody else also use admixture for their results.
and admixture doesn't really "look" at anything, it takes your raw data and does a couple of very complicated calculations on it to arrive at the results.
it's not something simple like determining haplogroups or finding common snps or centimorgans between 2 individuals.


It sure won’t be the Russian you mention.
it won't be.. and it shouldn't be, because that's not the point of admixture.

[Zoro]

23andme, ancestry and everybody else also use admixture for their results.
and admixture doesn't really "look" at anything, it takes your raw data and does a couple of very complicated calculations on it to arrive at the results.

Are you for real man? It doesn’t look at anything

[vbnetkhio]

Are you for real man? It doesn’t look at anything

http://software.genetics.ucla.edu/admixture/


Specifically, ADMIXTURE uses a block relaxation approach to alternately update allele frequency and ancestry fraction parameters. Each block update is handled by solving a large number of independent convex optimization problems, which are tackled using a fast sequential quadratic programming algorithm. Convergence of the algorithm is accelerated using a novel quasi-Newton acceleration method. The algorithm outperforms EM algorithms and MCMC sampling methods by a wide margin. For details, see our publications.

as far as I can tell it doesn't look at individual snps. if you know something about block relaxation, sequential quadratic programming, and quasi-Newton acceleration, please explain it to us.

[Zoro]

http://software.genetics.ucla.edu/admixture/



as far as I can tell it doesn't look at individual snps. if you know something about block relaxation, sequential quadratic programming, and quasi-Newton acceleration, please explain it to us.

I asked Dilawer and he said all those things are techniques to speed up processing and the basis of admixture is Allele frequencies meaning that each SNP is looked at for the input samples and The SNP is assigned to a population based on allele frequencies

[SharpFork]

Eurogenes is completely reliable, weren't you that guy that claimed supposedly everyone was like East Eurasian or something? You push a lot of wacky theories and yet are baffled when something NOT designed for PRE-human populations doesn't model those samples exactly as you like, maybe because they don't use Neanderthal references? Maybe expecting Neanderthal to be model exactly through their descendants that have 2-5% of their admixture is something only an idiot would expect?

No no you are totally right and it's the entire calculator that's flawed, sure thing buddy.

[Zoro]

Eurogenes is completely reliable, weren't you that guy that claimed supposedly everyone was like East Eurasian or something? You push a lot of wacky theories and yet are baffled when something NOT designed for PRE-human populations doesn't model those samples exactly as you like, maybe because they don't use Neanderthal references? Maybe expecting Neanderthal to be model exactly through their descendants that have 2-5% of their admixture is something only an idiot would expect?

No no you are totally right and it's the entire calculator that's flawed, sure thing buddy.

Are you high on something? So you think I’m Dilawer? Only an idiot would believe Dilawer would waste his time here.

And yes a test that shows Altai Neanderthal as 97% African is flawed. Ok, I’ll put some of the blame on Admixture but you did notice that out of the Admixture calculators Eurogenes k13 made him the most African!

[firemonkey]

My father has a blue triangle with Siberian on chr 5
31 SNPS are listed

At the end it says chr5 30,850,106

I used target mode with Admix studio with my father's data . Did 30850206(got that slightly wrong) to 42000000



Lukasz K47

My father

Component %
West-Finnic 38.68%
East-Euro 18.86%
North-Sea_Germanic 12.31%
Nilotic 12.17%
Pamirian 10.13%
NW-Indian 5.19%
Papuan 1.36%
Volgan 1.31%


Oracle Calculator by www.dnagenics.com
v1.3 01/12/2019

Finished reading population data. 140 populations found.
47 components mode.

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


Mixed Mode:
1 66.02% Saami + 33.98% Estonian @ 25.616
2 76.95% Saami + 23.05% UKR_Lvov @ 26.015
3 70.70% Saami + 29.30% Russian @ 26.103
4 50.39% Estonian + 49.61% Saami @ 26.107
5 79.30% Saami + 20.70% Hungarian1 @ 26.460
6 86.33% Saami + 13.67% Masai @ 26.513
7 80.08% Saami + 19.92% Czech @ 26.526
8 80.08% Saami + 19.92% Croat @ 26.538
9 88.67% Saami + 11.33% Bulala @ 26.551
10 81.64% Saami + 18.36% Romanian_Apuseni @ 26.577

Least-squares method.

Using 1 populations approximation
1 100% Saami @ 27.672
2 100% Estonian @ 32.689
3 100% Russian @ 34.482
4 100% Hungarian1 @ 40.303
5 100% Aleut_Tlingit @ 40.345
6 100% Aleut @ 40.700
7 100% Hungarian2 @ 41.140
8 100% UKR_Lvov @ 41.287
9 100% Czech @ 41.290
10 100% Croat @ 41.758

Using 2 populations approximation
1 50% Saami + 50% Estonian @ 26.084
2 50% Saami + 50% Russian @ 26.942
3 50% Saami + 50% Saami @ 27.672
4 50% Saami + 50% UKR_Lvov @ 28.368
5 50% Saami + 50% Hungarian1 @ 28.713
6 50% Saami + 50% Czech @ 29.065
7 50% Saami + 50% Croat @ 29.207
8 50% Saami + 50% Hungarian2 @ 29.277
9 50% Saami + 50% Romanian_Apuseni @ 29.559
10 50% Saami + 50% Romania_Gorj @ 30.086

Using 3 populations approximation
1 33% Saami + 33% Saami + 33% Estonian @ 25.617
2 33% Saami + 33% Saami + 33% Russian @ 26.140
3 33% Saami + 33% Saami + 33% UKR_Lvov @ 26.384
4 33% Saami + 33% Saami + 33% Hungarian1 @ 26.880
5 33% Saami + 33% Saami + 33% Czech @ 27.049
6 33% Saami + 33% Saami + 33% Croat @ 27.105
7 33% Saami + 33% Saami + 33% Romanian_Apuseni @ 27.258
8 33% Saami + 33% Saami + 33% Hungarian2 @ 27.277
9 33% Saami + 33% Estonian + 33% Estonian @ 27.516
10 33% Saami + 33% Saami + 33% Romania_Gorj @ 27.543

Using 4 populations approximation
1 25% Saami + 25% Saami + 25% Saami + 25% Estonian @ 25.766
2 25% Saami + 25% Saami + 25% Saami + 25% UKR_Lvov @ 26.032
3 25% Saami + 25% Saami + 25% Estonian + 25% Estonian @ 26.084
4 25% Saami + 25% Saami + 25% Saami + 25% Russian @ 26.133
5 25% Saami + 25% Saami + 25% Estonian + 25% Russian @ 26.436
6 25% Saami + 25% Saami + 25% Saami + 25% Hungarian1 @ 26.504
7 25% Saami + 25% Saami + 25% Estonian + 25% Masai @ 26.548
8 25% Saami + 25% Saami + 25% Saami + 25% Czech @ 26.601
9 25% Saami + 25% Saami + 25% Saami + 25% Croat @ 26.627
10 25% Saami + 25% Saami + 25% Saami + 25% Romanian_Apuseni @ 26.705


nMonte Calculator by www.dnagenics.com
v1.1 15/12/2019

Finished reading population data. 140 populations found.
47 components mode.

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

Distance: 26.27

63.60% Saami
29.00% Estonian
04.40% Bulala
03.00% Masai



MDLP world 22


Component %
North-East-European 43.31%
Indo-Iranian 19.03%
North-European-Mesolithic 16.91%
Atlantic_Mediterranean_Neolithic 11.49%
South-America_Amerind 3.96%
Mesoamerican 3.54%
East-South-Asian 1.48%
Austronesian 0.28%


Oracle Calculator by www.dnagenics.com
v1.3 01/12/2019

Finished reading population data. 275 populations found.
22 components mode.

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


Mixed Mode:
1 80.08% Finnish + 19.92% Proto-Indo-Iranian @ 6.666
2 76.95% Finnish + 23.05% Kalash @ 6.705
3 79.30% Inkeri + 20.70% Proto-Indo-Iranian @ 7.101
4 76.17% Inkeri + 23.83% Kalash @ 7.296
5 76.17% Karelian + 23.83% Kalash @ 8.323
6 78.52% Karelian + 21.48% Proto-Indo-Iranian @ 8.517
7 76.95% Finnish-South + 23.05% Kalash @ 9.182
8 79.30% Finnish-South + 20.70% Proto-Indo-Iranian @ 9.366
9 74.61% Vepsa + 25.39% Kalash @ 10.297
10 77.73% Vepsa + 22.27% Proto-Indo-Iranian @ 10.736

Least-squares method.

Using 1 populations approximation
1 100% Finnish @ 24.789
2 100% Inkeri @ 25.991
3 100% Finnish-South @ 26.073
4 100% Karelian @ 26.574
5 100% Tatar_Kryashen @ 26.870
6 100% Aleut @ 27.697
7 100% Tartar_Mishar @ 27.932
8 100% Tatar @ 27.974
9 100% Swedish_V @ 28.133
10 100% Komi @ 28.396

Using 2 populations approximation
1 50% Finnish-North + 50% Colville @ 21.279
2 50% Vepsa + 50% Miwok @ 21.325
3 50% Finnish-South + 50% Tatar @ 21.330
4 50% Tatar + 50% Finnish-North @ 21.416
5 50% Finnish-South + 50% Colville @ 21.542
6 50% Finnish-North + 50% Tatar_Lithuania @ 21.624
7 50% Finnish-South + 50% Tatar_Lithuania @ 21.642
8 50% Tatar_Kryashen + 50% Finnish-North @ 21.752
9 50% Finnish-South + 50% Aleut @ 21.774
10 50% Karelian + 50% Miwok @ 21.775

Using 3 populations approximation
1 33% Finnish-South + 33% Karelian + 33% Miwok @ 19.423
2 33% Karelian + 33% Karelian + 33% Miwok @ 19.490
3 33% Russian_North + 33% Finnish-North + 33% Miwok @ 19.534
4 33% Finnish + 33% Karelian + 33% Miwok @ 19.539
5 33% Finnish + 33% Finnish-South + 33% Miwok @ 19.617
6 33% Finnish-North + 33% Mordovian + 33% Miwok @ 19.633
7 33% Finnish + 33% Finnish + 33% Miwok @ 19.705
8 33% Vepsa + 33% Vepsa + 33% Miwok @ 19.796
9 33% Inkeri + 33% Finnish-South + 33% Miwok @ 19.804
10 33% Inkeri + 33% Karelian + 33% Miwok @ 19.805

Using 4 populations approximation
1 25% Finnish-North + 25% Finnish-North + 25% Mordovian + 25% Miwok @ 19.296
2 25% Finnish-South + 25% Finnish-South + 25% Tatar_Kryashen + 25% Miwok @ 19.412
3 25% Vepsa + 25% Vepsa + 25% Norwegian_V + 25% Ste7 @ 19.557
4 25% Finnish + 25% Finnish + 25% Finnish-South + 25% Miwok @ 19.642
5 25% Finnish + 25% Finnish + 25% Finnish + 25% Miwok @ 19.672
6 25% Finnish-North + 25% Finnish-North + 25% Mordovian + 25% Mexican @ 19.683
7 25% Finnish + 25% Finnish + 25% Karelian + 25% Miwok @ 19.692
8 25% Finnish-North + 25% Finnish-North + 25% German-North + 25% Ste7 @ 19.705
9 25% Finnish-South + 25% Finnish-South + 25% Karelian + 25% Miwok @ 19.770
10 25% Finnish + 25% Finnish + 25% Finnish + 25% Ste7 @ 19.772

nMonte Calculator by www.dnagenics.com
v1.1 15/12/2019

Finished reading population data. 275 populations found.
22 components mode.

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

Distance: 10.49

66.80% Finnish
15.80% Proto-Indo-Iranian
04.40% Colombian
04.00% Colville
03.00% Finnish-South
02.40% Swedish_V
01.40% Kalash
01.40% Surui
00.80% Inkeri


My father

Dodecad 12b

Component %
North_European 72.15%
South_Asian 8.94%
East_African 7.30%
Atlantic_Med 4.69%
East_Asian 3.42%
Gedrosia 2.73%
Caucasus 0.77%


Oracle Calculator by www.dnagenics.com
v1.3 01/12/2019

Finished reading population data. 223 populations found.
12 components mode.

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


Mixed Mode:
1 89.45% Lithuanians + 10.55% Chamar_M @ 13.194
2 89.45% Lithuanians + 10.55% Hakkipikki_M @ 13.215
3 88.67% Lithuanians + 11.33% Kol_M @ 13.232
4 89.45% Lithuanians + 10.55% SAKILLI @ 13.233
5 89.45% Lithuanians + 10.55% Chenchus_M @ 13.247
6 89.45% Lithuanians + 10.55% MALAYAN @ 13.273
7 88.67% Lithuanians + 11.33% Dusadh_M @ 13.283
8 88.67% Lithuanians + 11.33% Kanjars_M @ 13.284
9 88.67% Lithuanians + 11.33% Uttar_Pradesh_Scheduled_Caste_M @ 13.287
10 89.45% Lithuanians + 10.55% Gond_M @ 13.292

Least-squares method.

Using 1 populations approximation
1 100% FIN30 @ 16.840
2 100% Finnish_D @ 16.920
3 100% Lithuanians @ 17.594
4 100% Lithuanian_D @ 17.809
5 100% Russian @ 18.735
6 100% Russian_D @ 19.609
7 100% Mordovians_Y @ 20.069
8 100% Russian_B @ 20.601
9 100% Belorussian @ 21.295
10 100% Mixed_Slav_D @ 22.353

Using 2 populations approximation
1 50% FIN30 + 50% Lithuanian_D @ 16.363
2 50% FIN30 + 50% Lithuanians @ 16.384
3 50% Finnish_D + 50% Lithuanian_D @ 16.430
4 50% Finnish_D + 50% Mordovians_Y @ 16.537
5 50% Finnish_D + 50% Lithuanians @ 16.557
6 50% Finnish_D + 50% Russian @ 16.609
7 50% Lithuanians + 50% Russian @ 16.735
8 50% Finnish_D + 50% Russian_D @ 16.819
9 50% FIN30 + 50% Mordovians_Y @ 16.825
10 50% FIN30 + 50% Finnish_D @ 16.839

Using 3 populations approximation
1 33% Finnish_D + 33% Finnish_D + 33% Mordovians_Y @ 16.185
2 33% Finnish_D + 33% Lithuanians + 33% Mordovians_Y @ 16.195
3 33% Finnish_D + 33% Lithuanians + 33% Russian @ 16.239
4 33% FIN30 + 33% Lithuanians + 33% Mordovians_Y @ 16.257
5 33% FIN30 + 33% Lithuanians + 33% Russian @ 16.279
6 33% FIN30 + 33% Finnish_D + 33% Mordovians_Y @ 16.288
7 33% FIN30 + 33% FIN30 + 33% Lithuanian_D @ 16.302
8 33% FIN30 + 33% Finnish_D + 33% Lithuanian_D @ 16.323
9 33% FIN30 + 33% FIN30 + 33% Lithuanians @ 16.347
10 33% Finnish_D + 33% Finnish_D + 33% Lithuanian_D @ 16.381

Using 4 populations approximation
1 25% Finnish_D + 25% Finnish_D + 25% Lithuanians + 25% Mordovians_Y @ 16.056
2 25% FIN30 + 25% FIN30 + 25% Lithuanians + 25% Mordovians_Y @ 16.117
3 25% Finnish_D + 25% Finnish_D + 25% Lithuanian_D + 25% Mordovians_Y @ 16.168
4 25% Finnish_D + 25% Finnish_D + 25% Finnish_D + 25% Mordovians_Y @ 16.190
5 25% Finnish_D + 25% Finnish_D + 25% Lithuanians + 25% Russian @ 16.200
6 25% FIN30 + 25% FIN30 + 25% Lithuanians + 25% Russian @ 16.234
7 25% Finnish_D + 25% Finnish_D + 25% Lithuanian_D + 25% Russian @ 16.269
8 25% FIN30 + 25% FIN30 + 25% Lithuanian_D + 25% Mordovians_Y @ 16.282
9 25% FIN30 + 25% FIN30 + 25% Finnish_D + 25% Mordovians_Y @ 16.296
10 25% FIN30 + 25% FIN30 + 25% Lithuanians + 25% Lithuanian_D @ 16.341


nMonte Calculator by www.dnagenics.com
v1.1 15/12/2019

Finished reading population data. 223 populations found.
12 components mode.

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

Distance: 13.64

47.40% Lithuanians
42.00% Finnish_D
03.20% Chamar_M
03.20% MALAYAN
01.80% Sandawe_He
01.80% Somali_D
00.40% Kanjars_M
00.20% Kol_M


HarappaWorld

Component %
NE-Euro 64.11%
Mediterranean 10.89%
E-African 9.99%
S-Indian 9.49%
Baloch 2.82%
NE-Asian 1.22%
Caucasian 1.03%
American 0.46%


Oracle Calculator by www.dnagenics.com
v1.3 01/12/2019

Finished reading population data. 384 populations found.
16 components mode.

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


Mixed Mode:
1 88.67% lithuanian + 11.33% gumuz @ 10.877
2 88.67% lithuanian + 11.33% anuak @ 10.954
3 87.89% lithuanian + 12.11% sudanese @ 10.986
4 87.89% lithuanian + 12.11% ariblacksmith @ 11.085
5 87.11% lithuanian + 12.89% maasai @ 11.148
6 87.89% lithuanian + 12.11% aricultivator @ 11.148
7 87.11% lithuanian + 12.89% maasai @ 11.209
8 86.33% lithuanian + 13.67% hadza @ 11.324
9 88.67% lithuanian + 11.33% paniya @ 11.432
10 87.89% lithuanian + 12.11% bhunjia @ 11.636

Least-squares method.

Using 1 populations approximation
1 100% russian @ 16.443
2 100% belorussian @ 17.173
3 100% lithuanian @ 17.192
4 100% mordovian @ 17.810
5 100% russian @ 17.933
6 100% ukranian @ 20.046
7 100% finnish @ 22.887
8 100% slovenian @ 26.502
9 100% n-european @ 27.162
10 100% chuvash @ 28.026

Using 2 populations approximation
1 50% finnish + 50% singapore-indian-d @ 14.916
2 50% finnish + 50% n-european @ 15.819
3 50% russian + 50% lithuanian @ 15.895
4 50% finnish + 50% slovenian @ 16.089
5 50% lithuanian + 50% mordovian @ 16.118
6 50% finnish + 50% utahn-white @ 16.136
7 50% ukranian + 50% finnish @ 16.208
8 50% russian + 50% belorussian @ 16.248
9 50% finnish + 50% hungarian @ 16.376
10 50% russian + 50% russian @ 16.443

Using 3 populations approximation
1 33% finnish + 33% finnish + 33% singapore-indian-d @ 12.540
2 33% russian + 33% finnish + 33% singapore-indian-d @ 13.214
3 33% lithuanian + 33% lithuanian + 33% singapore-indian-d @ 13.528
4 33% ukranian + 33% finnish + 33% singapore-indian-d @ 14.034
5 33% russian + 33% lithuanian + 33% singapore-indian-d @ 14.832
6 33% belorussian + 33% lithuanian + 33% singapore-indian-d @ 14.933
7 33% lithuanian + 33% russian + 33% singapore-indian-d @ 15.040
8 33% finnish + 33% n-european + 33% singapore-indian-d @ 15.126
9 33% finnish + 33% slovenian + 33% singapore-indian-d @ 15.662
10 33% lithuanian + 33% mordovian + 33% singapore-indian-d @ 15.793

Using 4 populations approximation
1 25% finnish + 25% finnish + 25% n-european + 25% singapore-indian-d @ 12.052
2 25% finnish + 25% finnish + 25% slovenian + 25% singapore-indian-d @ 12.605
3 25% lithuanian + 25% lithuanian + 25% lithuanian + 25% singapore-indian-d @ 12.873
4 25% russian + 25% russian + 25% finnish + 25% singapore-indian-d @ 13.454
5 25% lithuanian + 25% lithuanian + 25% russian + 25% singapore-indian-d @ 13.516
6 25% finnish + 25% finnish + 25% finnish + 25% singapore-indian-d @ 13.666
7 25% lithuanian + 25% lithuanian + 25% mordovian + 25% singapore-indian-d @ 13.884
8 25% russian + 25% russian + 25% lithuanian + 25% singapore-indian-d @ 14.288
9 25% finnish + 25% finnish + 25% finnish + 25% fiji-indian @ 14.478
10 25% belorussian + 25% belorussian + 25% lithuanian + 25% singapore-indian-d @ 14.512


nMonte Calculator by www.dnagenics.com
v1.1 15/12/2019

Finished reading population data. 384 populations found.
16 components mode.

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

Distance: 9.78

61.80% lithuanian
23.00% finnish
07.60% gumuz
07.20% paniya
00.20% anuak
00.20% pulliyar


No population oracle

Tolan K54

Component %
24-Finland_SW 78.96%
14-Canary 19.49%
12-Sicily_Italia 1.55%

[vbnetkhio]

I asked Dilawer and he said all those things are techniques to speed up processing and the basis of admixture is Allele frequencies meaning that each SNP is looked at for the input samples and The SNP is assigned to a population based on allele frequencies

if it was just that, the calculation would take a few seconds and there wouldn't be need for any of that.

[Zoro]

My father has a blue triangle with Siberian on chr 5
31 SNPS are listed

At the end it says chr5 30,850,106



I'm not sure what you're trying to say. Did SAPDA confirm what the other calculators show for your dad?

[Zoro]

if it was just that, the calculation would take a few seconds and there wouldn't be need for any of that.

How in the world would you know? Are you the programmer that created Admixture? Do you even know what the processes are to determine allele frequencies for each population and to clusterize them? If so please share that.

I'm not a programmer but I'm sure it is pretty time consuming. I would imagine first the program would determine the allele frequency for each SNP and figure out what populations those 1000s of samples to assign. For example if 100 out of 1000 samples have the right SNP then the program has to figure out how many of the 100 belong to population A, B, C or D and then assign that SNP to that population but it doesn't know beforehand which sample belongs to which population. Then it has to do it for each SNP. I'm sure that would take a long time.

[vbnetkhio]

How in the world would you know? Are you the programmer that created Admixture? Do you even know what the processes are to determine allele frequencies for each population and to clusterize them? If so please share that.

I'm not a programmer but I'm sure it is pretty time consuming. I would imagine first the program would determine the allele frequency for each SNP and figure out what populations those 1000s of samples to assign. For example if 100 out of 1000 samples have the right SNP then the program has to figure out how many of the 100 belong to population A, B, C or D and then assign that SNP to that population but it doesn't know beforehand which sample belongs to which population. Then it has to do it for each SNP. I'm sure that would take a long time.

are we talking about the making of a calculator, or calculating admixtures from a raw data file in e. g. eurogenes k13?

[firemonkey]

I'm not sure what you're trying to say. Did SAPDA confirm what the other calculators show for your dad?


I was using target mode with Admixture studio .As it was suggested that my father's Siberian on chromosome 5 was worth exploring further . Unfortunately the software seems to be buggy,after working well for a while it's now giving the same results irrespective of the chromosome used and start and end points .

[firemonkey]

Duplicate.

[SharpFork]

Are you high on something? So you think I’m Dilawer? Only an idiot would believe Dilawer would waste his time here.

And yes a test that shows Altai Neanderthal as 97% African is flawed. Ok, I’ll put some of the blame on Admixture but you did notice that out of the Admixture calculators Eurogenes k13 made him the most African!
First learn to read English and reading what people actually write and stop wasting our time.

Also start comprehending that you CANNOT expect a calculator made for ANATOMICALLY MODERN HUMANS, especially one made for post Ice Age europeans and thus non African population, to be able to model a population whose genome can't be even fully reconstructed through their remain in modern humans.

Let's make another example, let's say you have an Amerindian, an Australian Aboriginal, a Khoisan person, a Yoruba person and each of them is 5% Kurdish genetically and then you have an Azerbaijani person, do you seriously expect to be able to model a fully Kurdish person through a mix of all those populations and NOT through a pure Azerbaijani? Seriously?

[Zoro]

First learn to read English and reading what people actually write and stop wasting our time.

Also start comprehending that you CANNOT expect a calculator made for ANATOMICALLY MODERN HUMANS, especially one made for post Ice Age europeans and thus non African population, to be able to model a population whose genome can't be even fully reconstructed through their remain in modern humans.

Let's make another example, let's say you have an Amerindian, an Australian Aboriginal, a Khoisan person, a Yoruba person and each of them is 5% Kurdish genetically and then you have an Azerbaijani person, do you seriously expect to be able to model a fully Kurdish person through a mix of all those populations and NOT through a pure Azerbaijani? Seriously?

Dude if you’re goina give an analogy at least give one which makes sense. How the freak is your analogy equivalent.?

You’re too fixated on the fact that there was only 5 percent geneflow from Eurasians to Altai Neander and VISA VERSA. Guess what it doesn’t make a difference to Admixture program and here’s why:

Let’s say that you have a k2 calculator. One component is African the other Eurasian (doesn’t matter whether E. Or W. Eurasian). If the calculator is properly designed to disregard ancestral alleles shared by Africans, Eurasians, and Neander and only consider DERIVED alleles, the calculator should output 100% Eurasian for the Neander sample since the only options are Eurasian and African. Therefore that 5% is 100% Eurasian

There’s no way you’re goina win this since I got help from Dilawer on this answer and he actually makes software and freakin genotypes DNA sequences unlike the blogger we are talking about

[SharpFork]

Dude if you’re goina give an analogy at least give one which makes sense. How the freak is your analogy equivalent.?

You’re too fixated on the fact that there was only 5 percent geneflow from Eurasians to Altai Neander and VISA VERSA. Guess what it doesn’t make a difference to Admixture program and here’s why:

Let’s say that you have a k2 calculator. One component is African the other Eurasian (doesn’t matter whether E. Or W. Eurasian). If the calculator is properly designed to disregard ancestral alleles shared by Africans, Eurasians, and Neander and only consider DERIVED alleles, the calculator should output 100% Eurasian for the Neander sample since the only options are Eurasian and African. Therefore that 5% is 100% Eurasian

There’s no way you’re goina win this since I got help from Dilawer on this answer and he actually makes software and freakin genotypes DNA sequences unlike the blogger we are talking about
What if the remaining 95% of Neanderthal DNA is overall more similar to the one those peripherical African population do? Why don't you consider that? Not all calculators measure the same things, how would you properly measure descend through a ancestry based calculator anyway? Even your SAPDA calculator says Neanderthals are 30% African...

[Zoro]

What if the remaining 95% of Neanderthal DNA is overall more similar to the one those peripherical African population do? Why don't you consider that? Not all calculators measure the same things, how would you properly measure descend through a ancestry based calculator anyway? Even your SAPDA calculator says Neanderthals are 30% African...

Carefully read what I said. There are a few experienced calculator makers who can filter out ANCESTRAL alleles when making the calculator . Therefore there shouldn’t be any allele sharing between Neander and Africans when done properly

[SharpFork]

Carefully read what I said. There are a few experienced calculator makers who can filter out ANCESTRAL alleles when making the calculator . Therefore there shouldn’t be any allele sharing between Neander and Africans when done properly

And why should a calculator made for post-ice age Eurasians do any of that? Answer the question.

[Zoro]

And why should a calculator made for post-ice age Eurasians do any of that? Answer the question.

You’ll find clues in Dilawer’s statement


Eurasians, whether Europeans or Asians, are genetically very similar to each other, with genetic similarities far outweighing any differences they have. Most Eurasians have descended from a few thousand individuals who roamed Eurasia during the Upper Paleolithic and thus share a huge amount of DNA. Thus an admixture inference program needs to determine which mutations are shared due to distant common origins and which ones due to more recent introgression.

It is for this reason that some bioinformatic software such as qpDstat, qp3pop, and qpAdm, which are included in the ADMIXTOOLS software suite available at Reich Lab, use outgroups to filter out older common ancestral alleles.

An extreme example of how ancestral alleles and SNP ascertainment bias distort admixture calculations using the program ADMIXTURE is Neanderthal. When the Neanderthal is processed with an ADMIXTURE based calculator, regardless of calculator design, the results show it to be around 90% African. Of course we know this inference is incorrect because the Neanderthal lineage diverged from humans over 500,000 years ago, with some hybridization with humans occurring around 50,000 years ago in Eurasia and NOT in Africa.




That’s the reason serious researchers use qpAdm and not Admixture or G25 with nMonte. SAPDA uses the same principle. For example, steppe populations are high in EHG. Since Europeans have alot of ancestry from Corded Ware and the like Admixture or G25 will sometimes show Europeans closer to Iron Age steppe nomads than some west Asians because of all the EHG. But if you were to use EHG as an outgroup, then you’ll have a better idea which steppe population contributed more to Europeans.