(NEW) GenePlaza K25 and K29 Modern Calculator Results

[PaschalisB]

After the rerun I get 47.1% Southern Slavic, 28.5% Greek-Albanian and 17.3% Caucasian.

 

[Pax Augusta]

That's a little disconcerting, I thought he would have used legit samples. But I guess samples like "Dorkymon" is evident of the fact the used user uploads.

Yes, Jovialis. He used user uploads.

 

[Sile]

Italian_N could be an Italian user active on anthrogenica who lives in north Italy but he is 3/4 from south Italy. I'm saying this because I've seen he was the only Italian who sent his raw data to Kurd and in previous spreadsheet his results were labeled as Italian_N.

Who is this one North-italian ? ............

 

[Pax Augusta]

Who is this one North-italian ? ............

I've already answered on this, Sile.

 

[Jovialis]

So theoretically, any t-roll could have lied, and said they were any ethnicity, and would have made the cut for a reference population? Are the components based on users as well?

 

[Sile]

Thanks for sharing. There are fully northern Italians with somewhat similar results to yours in northern eastern Italy. Interestingly your German part doesn't show, and if you're half southern German, your Italian side seems to be more southern-shifted than Tuscans who are central Italian.




Parabiago and Castellanza are in Lombardy and not really in the Po Valley (especially Castellanza). Assuming your 50% German is not an outlier for southern Germany, Lombards are never more southern in this way than Italian (HGDP). Italian_hgdp is based on Italian_Bergamo and it's a sample from Lombardy. I have seen more than 50 results of people from Lombardy. On Harappa Lombards on average have 29/30/31% of NE component, I have a friend from near Castellanza who scores 32% of NE, and you have 36% NE. There are many Piedmontese and Venetians who score 36%. Of course I'm speaking of fully northern Italians, there was a huge migration in Lombardy and other parts of north Italy of people from southern Italy in the last 120 years.

If I compare mine below from harappa with Alex , there are some similarities ................I think Harappa is not a good test fro a european only a middle-eastener

Admix Results (sorted):

# Population Percent
1 Mediterranean 32.49
2 NE-Euro 32.28
3 Caucasian 20.13
4 Baloch 8.21
5 SW-Asian 5.89


Finished reading population data. 377 populations found.
16 components mode.

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

Least-squares method.

Using 1 population approximation:
1 italian_hgdp @ 6.817736
2 romanian-a_behar @ 11.724042
3 tuscan_hapmap @ 11.845555
4 tuscan_hgdp @ 12.231710
5 tuscan_1000genomes @ 12.757281
6 bulgarian_yunusbayev @ 13.483592
7 spaniard_behar @ 15.001562
8 serbian_harappa @ 15.199749
9 french_hgdp @ 15.544718
10 spaniard_1000genomes @ 16.260679
11 hungarian_behar @ 19.697365
12 slovenian_xing @ 21.887278
13 ashkenazi_harappa @ 22.299673
14 utahn-white_1000genomes @ 24.328009
15 n-european_xing @ 25.091066
16 puerto-rican_1000genomes @ 25.139376
17 ashkenazy-jew_behar @ 25.254951
18 british_1000genomes @ 25.866997
19 utahn-white_hapmap @ 25.944796
20 orcadian_hgdp @ 28.933249

Using 2 populations approximation:
1 50% french_hgdp +50% tuscan_1000genomes @ 2.632360


Using 3 populations approximation:
1 50% n-european_xing +25% sardinian_hgdp +25% turk_behar @ 1.993485

 

[davef]

So theoretically, any t-roll could have lied, and said they were any ethnicity, and would have made the cut for a reference population? Are the components based on users as well?

So they (t-rolls) tricked the guy (Kurd) with fraudulent claims of ethnicity and thus compromised the quality of his product! What could happen next is people complaining about their results and him having to resort to spending more time, more money fixing things he never never wouldve had to had people been honest about themselves!

This is detrimental to not just himself and his clients, it harms his business as well!

He's doing us a huge favor and people spit back.

 

[Jovialis]

So they tricked the guy (Kurd) with fraudulent claims of ethnicity and thus compromised the quality of his product! What could happen next is people complaining about their results and him having to resort to spending more time, more money fixing things he never never wouldve had to had people been honest about themselves!
This is detrimental to not just himself and his clients, it harms his business as well!
He's doing us a huge favor and people spit back.

I don't want to jump to conclusions. Because It may not be the case, but I would like to verify that first in regards to the components. In the case for the Italian_N reference sample; maybe it wasn't a t-roll in that regard, but it was a poor choice.

One thing is clear, Dibran gets 100% Greek-Albanian, because that's exactly what he is (Albanian). If someone's ethnicity doesn't have a dedicated component, they will get some random assortment of components, which may or may not make sense.

 

[Pax Augusta]

All is well with the world again. Lmao. Extremely Accurate. I suppose since my father shifts a bit with his east med he gets more Sicilian.

Lmao yes, it looks like you're the Greek-Albanian sample.

So theoretically, any t-roll could have lied, and said they were any ethnicity, and would have made the cut for a reference population? Are the components based on users as well?

If mistakes have been made, I hope they have been made in good faith.

But yes, if user uploads have been used for components and reference population, and someone could have lied and or given information not completely correct on own ethnicity... You know.

One thing is clear, Durban gets 100% Greek-Albanian, because that's exactly what he is (Albanian). If someone's ethnicity doesn't have a dedicated component, they will get some random assortment of components, which may or may not make sense.

That's extremely true.

 

[poi]

With the K29 update, I no longer have ANY EastAsian/SoutheastAsian/Siberian/Polar ... just 0%. This does not sound right given my ancestral location. I am getting consistent ~5% East/SoutheastAsian/Siberian/Polar in all other calculators. Only 23andme gives me 100% South Asian, but they probably mean the modern population within the last few centuries.

 

[poi]

All is well with the world again. Lmao. Extremely Accurate. I suppose since my father shifts a bit with his east med he gets more Sicilian.

You are getting 100% because you could be the reference population for your ethnic group. If so, this calculator is not useful for you.

 

[Pax Augusta]

If I compare mine below from harappa with Alex , there are some similarities ................I think Harappa is not a good test fro a european only a middle-eastener

Admix Results (sorted):

# Population Percent
1 Mediterranean 32.49
2 NE-Euro 32.28
3 Caucasian 20.13
4 Baloch 8.21
5 SW-Asian 5.89


Finished reading population data. 377 populations found.
16 components mode.

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

Least-squares method.

Using 1 population approximation:
1 italian_hgdp @ 6.817736
2 romanian-a_behar @ 11.724042
3 tuscan_hapmap @ 11.845555
4 tuscan_hgdp @ 12.231710
5 tuscan_1000genomes @ 12.757281
6 bulgarian_yunusbayev @ 13.483592
7 spaniard_behar @ 15.001562
8 serbian_harappa @ 15.199749
9 french_hgdp @ 15.544718
10 spaniard_1000genomes @ 16.260679
11 hungarian_behar @ 19.697365
12 slovenian_xing @ 21.887278
13 ashkenazi_harappa @ 22.299673
14 utahn-white_1000genomes @ 24.328009
15 n-european_xing @ 25.091066
16 puerto-rican_1000genomes @ 25.139376
17 ashkenazy-jew_behar @ 25.254951
18 british_1000genomes @ 25.866997
19 utahn-white_hapmap @ 25.944796
20 orcadian_hgdp @ 28.933249

Population percent is not tha bad, but I agree with you that the Harappa Oracles aren't that good.

Please, could you post also your wife Harappa's results? As I know, she is Venetian as you.

 

[Angela]

I don't want to jump to conclusions. Because It may not be the case, but I would like to verify that first in regards to the components. In the case for the Italian_N reference sample; maybe it wasn't a t-roll in that regard, but it was a poor choice.
One thing is clear, Dibran gets 100% Greek-Albanian, because that's exactly what he is (Albanian). If someone's ethnicity doesn't have a dedicated component, they will get some random assortment of components, which may or may not make sense.

That's exactly right.

I think the "dedicated" components or clusters are the vertical ones on your chart. I don't think he has a dedicated or reference Italian cluster of any sort. That's been the problem with his analysis from the beginning.

On Dodecad, for example, when there's a "Caucasus" cluster, as in K12b, for example, North Italians do get it, but it's a lot lower than that of mainland Greeks, and even of Bulgarians and Romanians. You'd never guess that looking at these results.

This is also why you have to do this with "regional" clusters from more than one country.

This is giving a totally distorted view of the genetics of a lot of Southern Europeans, and especially Italians, I'm sorry to say.

 

[Dibran]

With the K29 update, I no longer have ANY EastAsian/SoutheastAsian/Siberian/Polar ... just 0%. This does not sound right given my ancestral location. I am getting consistent ~5% East/SoutheastAsian/Siberian/Polar in all other calculators. Only 23andme gives me 100% South Asian, but they probably mean the modern population within the last few centuries.

Your last sentence I think lol.

 

[Dibran]

You are getting 100% because you could be the reference population for your ethnic group. If so, this calculator is not useful for you.

I am not though. Neither are my parents. Never sent Kurd anything, and per his reference display, my kits were not included. My father doesnt get 100. He hets 81 and 16 or so Sicilian, 3 east slavic. etc. My mom gets 100 Albanian. Before removing Bulgarian from south slavic, she was roughly 50/50. Also, they removed overlapping kits, and added Serbs to south Slavic and Kosova to Albanian.

 

[davef]

If I compare mine below from harappa with Alex , there are some similarities ................I think Harappa is not a good test fro a european only a middle-eastener

Admix Results (sorted):

# Population Percent
1 Mediterranean 32.49
2 NE-Euro 32.28
3 Caucasian 20.13
4 Baloch 8.21
5 SW-Asian 5.89


Finished reading population data. 377 populations found.
16 components mode.

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

Least-squares method.

Using 1 population approximation:
1 italian_hgdp @ 6.817736
2 romanian-a_behar @ 11.724042
3 tuscan_hapmap @ 11.845555
4 tuscan_hgdp @ 12.231710
5 tuscan_1000genomes @ 12.757281
6 bulgarian_yunusbayev @ 13.483592
7 spaniard_behar @ 15.001562
8 serbian_harappa @ 15.199749
9 french_hgdp @ 15.544718
10 spaniard_1000genomes @ 16.260679
11 hungarian_behar @ 19.697365
12 slovenian_xing @ 21.887278
13 ashkenazi_harappa @ 22.299673
14 utahn-white_1000genomes @ 24.328009
15 n-european_xing @ 25.091066
16 puerto-rican_1000genomes @ 25.139376
17 ashkenazy-jew_behar @ 25.254951
18 british_1000genomes @ 25.866997
19 utahn-white_hapmap @ 25.944796
20 orcadian_hgdp @ 28.933249

Using 2 populations approximation:
1 50% french_hgdp +50% tuscan_1000genomes @ 2.632360


Using 3 populations approximation:
1 50% n-european_xing +25% sardinian_hgdp +25% turk_behar @ 1.993485

Whats your eurogenes k15 like? Here's Alex's from a previous list of his:

https://m.imgur.com/sBsixpd

 

[Eblö]

What is "caucasus" exactly in this calculator?I got 15.2 as a central european slav,and even in my European, half was south slavic the other half baltic with zero western or eastern slav. It doesnt seem to make too much sense...

 

[Angela]

What is "caucasus" exactly in this calculator?I got 15.2 as a central european slav,and even in my European, half was south slavic the other half baltic with zero western or eastern slav. It doesnt seem to make too much sense...

Circassian, I believe, or Adyghei if you prefer.

 

[Ghani]

So theoretically, any t-roll could have lied, and said they were any ethnicity, and would have made the cut for a reference population? Are the components based on users as well?

The samples that he uses are project member results not references. The references are from public datasets like HGDP and so on. The references are all set to 100%

But you are right he has to rely on project members and customers giving their correct known ancestry, but of course some people don’t know what it is past a few generations ago

 

[Ghani]

With the K29 update, I no longer have ANY EastAsian/SoutheastAsian/Siberian/Polar ... just 0%. This does not sound right given my ancestral location. I am getting consistent ~5% East/SoutheastAsian/Siberian/Polar in all other calculators. Only 23andme gives me 100% South Asian, but they probably mean the modern population within the last few centuries.

Actually realistically unless you are from SE Asia or Siberia you didn’t get your Siberian or SE Asian from those groups directly anyway. You probably got it from one of those 5 Turkic groups or Tajikistanis that he has in the calculator. So if that’s the case then the calculator is more useful for you than the other ones