Page 2 of 5 FirstFirst 1234 ... LastLast
Results 11 to 20 of 48

Thread: Has G25 been discontinued?

  1. #11
    Registered Users
    Posts
    284
    Sex
    Location
    Germany
    Ethnicity
    German
    Y-DNA (P)
    R1b (> R-CTS4528)
    mtDNA (M)
    U5b2b3

    Germany Imperial
    Beside possible private reasons only for making a pause I think Davidski did realise some weaknesses of G25. I can mention a particular weakness. G25 kan hardly distinguish between a Germanic-Slavic-like genetic mixture and a Germanic-Finnic-like genetic mixture. So I as an Eastern German always get high Sweden similarities, which of course is genetically wrong. In terms of K36 there is no such problem. Davidski got somewhat annoyed by complaints about G25 not being correct (not from me) and these complaints were motivated with examples from the modern data. So as you all probably know Davidski once told this and announced to perspectively remove the modern spread sheets and told G25 is most suitable for ancient DNA only.

    But I don't believe the latter. G25 will have the same weaknesses for ancient DNA too, but nobody can prove it that simple as with G25 coords for modern people. And knowing this might well be somewhat demotivating, because there is no satisfaction anymore when you've lost the conviction that coords are very good for that purpose. In fact it must be dissatisfactory to know the own "open source" K36 data and coords are even more suitable. (This is my opinion.)

    Re-calculating all ancient DNA within K36 or another new system is horrible much work. And just quitting G25 without replacing it with something is dissatisfactory as well. So this is what Davidski well might be into right now.
    Last edited by rothaer; 06-22-2021 at 05:48 AM.

  2. The Following 8 Users Say Thank You to rothaer For This Useful Post:

     ChrisR (07-14-2021),  digital_noise (06-21-2021),  dosas (06-23-2021),  FionnSneachta (06-24-2021),  MatAust21 (06-22-2021),  Nino90 (06-22-2021),  Tolan (06-23-2021),  Tomenable (06-22-2021)

  3. #12
    Quote Originally Posted by rothaer View Post
    Beside possible private reasons only for making a pause I think Davidski did realise some weaknesses of G25. I can mention a particular weakness. G25 kan hardly distinguish between a Germanic-Slavic-like genetic mixture and a Germanic-Finnic-like genetic mixture.
    Horses for courses. This is not a weakness of G25 or its underlying SmartPCA foundation . It's working exactly as you would expect a Global PCA to work. The granularity you're looking for is certainly possible using the same techniques, but it would no longer be a Global PCA but rather a European-centric one.

    Eurogenes K36, on the other hand, uses a fixed set of populations to determine ancestry proportions. Two of these populations are explicitly derived from Finnish and Eastern-European sources, so they happen to match your requirements perfectly.

    This does not make K36 a better tool for the vast majority of users, but it will definitely get you the granularity you're looking for in your specific use case. The Global 25 dataset and it's associated nMonte modeling might not be absolutely cutting edge any more, but they remain among the best available resources for ancestry analysis. Certainly the best available to the general public.

    Admixture calculators, such as K36, also still have a place. Particularly when they are built using references that are relevant to your ancestry. Some of the techniques that Global 25 helped popularize, such as nMonte modeling/Monte Carlo simulations, are applicable to admixture calculators as well.

    As a bonus, nMonte modeling with Admixture coordinates in Genoplot also gives you access to Oracles for calculators that may not have them built in.

    Explore your Genealogy

  4. The Following 6 Users Say Thank You to GenoPlot For This Useful Post:

     chocoholic (06-22-2021),  ChrisR (07-14-2021),  JMcB (06-24-2021),  K33 (06-24-2021),  Nino90 (06-23-2021),  rothaer (06-24-2021)

  5. #13
    Registered Users
    Posts
    969
    Sex

    Quote Originally Posted by rothaer View Post
    Beside possible private reasons only for making a pause I think Davidski did realise some weaknesses of G25. I can mention a particular weakness. G25 kan hardly distinguish between a Germanic-Slavic-like genetic mixture and a Germanic-Finnic-like genetic mixture. So I as an Eastern German always get high Sweden similarities, which of course is genetically wrong. In terms of K36 there is no such problem. Davidski got somewhat annoyed by complaints about G25 not being correct (not from me) and these complaints were motivated with examples from the modern data. So as you all probably know Davidski once told this and announced to perspectively remove the modern spread sheets and told G25 is most suitable for ancient DNA only.

    But I don't believe the latter. G25 will have the same weaknesses for ancient DNA too, but nobody can prove it that simple as with G25 coords for modern people. And knowing this might well be somewhat demotivating, because there is no satisfaction anymore when you've lost the conviction that coords are very good for that purpose. In fact it must be dissatisfactory to know the own "open source" K36 data and coords are even more suitable. (This is my opinion.)

    Re-calculating all ancient DNA within K36 or another new system is horrible much work. And just quitting G25 without replacing it with something is dissatisfactory as well. So this is what Davidski well might be into right now.
    I agree that Global25 may have issues, but the proximity to Sweden is not entirely coincidental, if I'm not mistaken Sweden had recent Balto-Slavic input as did East Germany. And in some parts of East Germany there may be real and recent Swedish ancestry. Right? Of all the Eurogenes calculators, K36 is perhaps the worst, the most preferred ones seem to be Eurogenes K13 and Dodecad K12b. More than a problem with Global25 and its operation, the real problem might be the samples that form the averages, which would not be accurate. Unfortunately it is less "reachable" than GEDmatch.
    [1] "distance%=2.9299"

    Ajeje Brazorf

    EEF,51.4
    BALTIC_BA,12.6
    GANJ_DAREH,10.8
    CHG,10.4
    MIDDLE_EAST,10
    ANE,2.4
    TAFORALT,1.8
    WHG,0.6

  6. The Following User Says Thank You to Ajeje Brazorf For This Useful Post:

     Nino90 (06-22-2021)

  7. #14
    Registered Users
    Posts
    202
    Sex
    Location
    Harappa
    Y-DNA (P)
    R-Y6
    mtDNA (M)
    W

    I sent Davidski an inquiry about G25 recently, and he got back to me yesterday confirming it'll be returning sometime next month:

    Screenshot 2021-06-22 9.34.55 PM.png

  8. The Following 7 Users Say Thank You to desi For This Useful Post:

     antpet (06-23-2021),  ChrisR (07-14-2021),  FrostAssassin0701 (06-23-2021),  JMcB (06-24-2021),  Nino90 (06-23-2021),  Sam1989 (06-23-2021),  Sapporo (06-23-2021)

  9. #15
    Registered Users
    Posts
    2,501
    Sex
    Location
    Sweden
    Ethnicity
    Italo-Germanic
    Nationality
    Swedish
    Y-DNA (P)
    R-L2 / R1b-U152
    mtDNA (M)
    H1a1

    Sweden Italy Italy 1861-1946 Sami Vatican Italy Tuscany
    Quote Originally Posted by Ajeje Brazorf View Post
    I agree that Global25 may have issues, but the proximity to Sweden is not entirely coincidental, if I'm not mistaken Sweden had recent Balto-Slavic input as did East Germany. And in some parts of East Germany there may be real and recent Swedish ancestry. Right? Of all the Eurogenes calculators, K36 is perhaps the worst, the most preferred ones seem to be Eurogenes K13 and Dodecad K12b. More than a problem with Global25 and its operation, the real problem might be the samples that form the averages, which would not be accurate. Unfortunately it is less "reachable" than GEDmatch.
    Swedes got Baltic admixture even in the early "viking age".

    Danes often lack this component. I also head that central and north Swedes on average have about 10-20% Finnish ancestry as well.


     
    Target: Swedish:Sweden9
    Distance: 1.1770% / 0.01177041
    86.4 Danish
    13.6 Estonian
    Target: Swedish:Sweden8
    Distance: 2.8251% / 0.02825060
    62.8 Danish
    19.2 Finnish
    18.0 Latvian
    Target: Swedish:Sweden7
    Distance: 1.8147% / 0.01814732
    86.4 Danish
    7.0 Estonian
    6.6 Finnish
    Target: Swedish:Sweden6
    Distance: 2.2456% / 0.02245604
    85.8 Danish
    12.8 Estonian
    1.4 Finnish
    Target: Swedish:Sweden5
    Distance: 2.1284% / 0.02128427
    94.0 Danish
    6.0 Latvian
    Target: Swedish:Sweden4
    Distance: 1.8942% / 0.01894164
    100.0 Danish
    Target: Swedish:Sweden3
    Distance: 2.1754% / 0.02175427
    80.6 Danish
    19.4 Finnish
    Target: Swedish:Sweden22
    Distance: 2.8216% / 0.02821608
    67.8 Danish
    21.0 Finnish
    11.2 Finnish_East
    Target: Swedish:Sweden21
    Distance: 2.8280% / 0.02828044
    75.2 Danish
    24.8 Finnish
    Target: Swedish:Sweden20
    Distance: 2.8324% / 0.02832414
    88.4 Danish
    6.6 Finnish_East
    5.0 Finnish
    Target: Swedish:Sweden2
    Distance: 2.5084% / 0.02508446
    83.0 Danish
    10.4 Finnish_East
    6.6 Estonian
    Target: Swedish:Sweden19
    Distance: 2.7925% / 0.02792487
    70.0 Danish
    30.0 Finnish
    Target: Swedish:Sweden18
    Distance: 3.3802% / 0.03380219
    94.4 Danish
    3.2 Finnish_East
    2.4 Estonian
    Target: Swedish:Sweden17
    Distance: 3.7043% / 0.03704275
    91.2 Danish
    8.8 Finnish_East
    Target: Swedish:Sweden16
    Distance: 2.0856% / 0.02085648
    84.2 Danish
    9.4 Finnish_East
    4.2 Latvian
    2.2 Finnish
    Target: Swedish:Sweden15
    Distance: 2.6598% / 0.02659808
    54.8 Danish
    25.4 Finnish
    19.8 Latvian
    Target: Swedish:Sweden13
    Distance: 1.9011% / 0.01901124
    87.4 Danish
    8.4 Latvian
    4.2 Finnish_East
    Target: Swedish:Sweden12
    Distance: 2.6627% / 0.02662668
    71.4 Danish
    16.6 Finnish_East
    12.0 Latvian
    Target: Swedish:Sweden11
    Distance: 2.5610% / 0.02561046
    70.4 Danish
    21.8 Estonian
    7.8 Finnish_East
    Target: Swedish:Sweden10
    Distance: 2.0903% / 0.02090263
    95.6 Danish
    4.4 Finnish_East
    Target: Swedish:Sweden1
    Distance: 1.9511% / 0.01951103
    85.4 Danish
    11.8 Finnish
    2.8 Latvian



    Vahaduo_ Global 25 Views (13).png
    Last edited by Nino90; 06-23-2021 at 06:27 PM.
    Hidden Content


    Target: Nino_scaled
    Distance: 2.3124% / 0.02312401 | R3P
    61.4 Swedish
    35.8 Spanish_Soria
    2.8 Nganassan

  10. The Following User Says Thank You to Nino90 For This Useful Post:

     JerryS. (06-24-2021)

  11. #16
    Gold Class Member
    Posts
    468
    Sex
    Y-DNA (P)
    R-Y33
    mtDNA (M)
    J1c2
    Y-DNA (M)
    E-Y6938
    mtDNA (P)
    G2a

    Russian Federation Russia Tatarstan Arms Russia Mordovia Timurid
    @Nino90

    I think all Scandinavians have this Balto-Slavic admixture in varying degrees. For some Swedes and Norwegians, there's also Uralic ancestry that's most prominent in the North.

    Comparing modern Scandinavians to IA Scandinavians, there's an increased affinity towards Balto-Slavic samples in the former, it's most evident in Swedes, but it's definitely not absent in Norwegians and Danes.

    For example, when comparing modern Norwegians with Norway_IA and other ancient Germanic samples, you see a shift towards Balto-Slavic ancestry represented here by AV2/Avar_Szolad2 sample.


     



    The shift in Norwegian ancestry towards Uralic, represented by kra001/Krasnoyarsk_BA, not as strong but still visible.

     




    I don't have modern Danish genotypes in my dataset, but you could very crudely model most of them on G25 as a mix of Denmark_IA and early-Slavic samples.

     

    YDNA (P): R-Y33
    YDNA (P, maternal line): R-Y20756
    YDNA(M): E-Y6938

  12. The Following 3 Users Say Thank You to altvred For This Useful Post:

     ChrisR (07-14-2021),  Nino90 (06-24-2021),  Tomenable (06-24-2021)

  13. #17
    Registered Users
    Posts
    2,501
    Sex
    Location
    Sweden
    Ethnicity
    Italo-Germanic
    Nationality
    Swedish
    Y-DNA (P)
    R-L2 / R1b-U152
    mtDNA (M)
    H1a1

    Sweden Italy Italy 1861-1946 Sami Vatican Italy Tuscany
    Quote Originally Posted by altvred View Post
    @Nino90

    I think all Scandinavians have this Balto-Slavic admixture in varying degrees. For some Swedes and Norwegians, there's also Uralic ancestry that's most prominent in the North.

    Comparing modern Scandinavians to IA Scandinavians, there's an increased affinity towards Balto-Slavic samples in the former, it's most evident in Swedes, but it's definitely not absent in Norwegians and Danes.

    For example, when comparing modern Norwegians with Norway_IA and other ancient Germanic samples, you see a shift towards Balto-Slavic ancestry represented here by AV2/Avar_Szolad2 sample.


     



    The shift in Norwegian ancestry towards Uralic, represented by kra001/Krasnoyarsk_BA, not as strong but still visible.

     




    I don't have modern Danish genotypes in my dataset, but you could very crudely model most of them on G25 as a mix of Denmark_IA and early-Slavic samples.

     

    Thank you for sharing.

    One of those danes got 30% Balto-Slavic admixture. That is quite a lot.

    I thought Danes were some kind of proxy for "clean" unmixed Germanic ancestry. But apparently I was wrong.¨

    What about Dutch or North Germans? Do they also have Balto-Slavic admixture?
    Hidden Content


    Target: Nino_scaled
    Distance: 2.3124% / 0.02312401 | R3P
    61.4 Swedish
    35.8 Spanish_Soria
    2.8 Nganassan

  14. #18
    Gold Class Member
    Posts
    468
    Sex
    Y-DNA (P)
    R-Y33
    mtDNA (M)
    J1c2
    Y-DNA (M)
    E-Y6938
    mtDNA (P)
    G2a

    Russian Federation Russia Tatarstan Arms Russia Mordovia Timurid
    Quote Originally Posted by Nino90 View Post
    Thank you for sharing.

    One of those danes got 30% Balto-Slavic admixture. That is quite a lot.

    I thought Danes were some kind of proxy for "clean" unmixed Germanic ancestry. But apparently I was wrong.¨

    What about Dutch or North Germans? Do they also have Balto-Slavic admixture?
    Nowadays, there's no such thing as an unmixed or pure population. Danes compared to Swedes were more isolated because they didn't have an Empire spanning Finland and the Baltics until the 18th century but you still had gene flow from neighboring populations.

    You mentioned that the Viking Age samples are already mixed, and it's the same situation with the ones from Denmark. A few of the VA individuals from Jutland, Funen, and Langeland appear to be unmixed Polabian Slavs with whom the Danes had historically attested contacts with I believe. So it's no surprise you got on average around 15-20% Balto-Slavic admixture in modern Danes.

     



    Unfortunately, the German samples publicly available aren't subdivided regionally (except the one's marked East_German on G25) so I can't know which one's are North Germans - but again, given the geographical proximity to Polabian/Pomeranian Slavs, I'd expect admixture in the range of what see in Danes.

    Outliers aside, there isn't a significant Balto-Slavic admix in the Dutch and nearby West Germans, but they are mixed with the pre-Germanic people who lived in the area (Celts/Gauls).

    I highly doubt it's possible to find a 'pure' proto-Germanic autosomal profile nowadays. We still see guys like that during the fall of the Roman Empire/Migration period, but it looks like in the Middle-Ages Germanic speakers were already mixed with surrounding and native pre-Germanic populations.
    YDNA (P): R-Y33
    YDNA (P, maternal line): R-Y20756
    YDNA(M): E-Y6938

  15. The Following 5 Users Say Thank You to altvred For This Useful Post:

     Aroon1916 (06-24-2021),  ChrisR (07-14-2021),  Coldmountains (06-24-2021),  Nino90 (06-24-2021),  rober_tce (06-25-2021)

  16. #19
    Moderator
    Posts
    2,594
    Sex

    Quote Originally Posted by Nino90 View Post
    Thank you for sharing.

    One of those danes got 30% Balto-Slavic admixture. That is quite a lot.

    I thought Danes were some kind of proxy for "clean" unmixed Germanic ancestry. But apparently I was wrong.¨

    What about Dutch or North Germans? Do they also have Balto-Slavic admixture?
    North Germans should have way more Slavic admix than Danes (medieval Slavic settlements up to Lübeck, internal german migration, refugees from Prussia/historical East Germany after the second world war). After the second world war, up to 30% of the population of Schleswig-Holstein and Niedersachsen were refugees mainly from the former eastern territories of Germany.

  17. The Following 5 Users Say Thank You to Coldmountains For This Useful Post:

     Alain (06-24-2021),  Aroon1916 (06-24-2021),  JerryS. (06-24-2021),  Nino90 (06-24-2021),  rothaer (06-24-2021)

  18. #20
    Registered Users
    Posts
    2,501
    Sex
    Location
    Sweden
    Ethnicity
    Italo-Germanic
    Nationality
    Swedish
    Y-DNA (P)
    R-L2 / R1b-U152
    mtDNA (M)
    H1a1

    Sweden Italy Italy 1861-1946 Sami Vatican Italy Tuscany
    Quote Originally Posted by altvred View Post
    Nowadays, there's no such thing as an unmixed or pure population. Danes compared to Swedes were more isolated because they didn't have an Empire spanning Finland and the Baltics until the 18th century but you still had gene flow from neighboring populations.

    You mentioned that the Viking Age samples are already mixed, and it's the same situation with the ones from Denmark. A few of the VA individuals from Jutland, Funen, and Langeland appear to be unmixed Polabian Slavs with whom the Danes had historically attested contacts with I believe. So it's no surprise you got on average around 15-20% Balto-Slavic admixture in modern Danes.

     



    Unfortunately, the German samples publicly available aren't subdivided regionally (except the one's marked East_German on G25) so I can't know which one's are North Germans - but again, given the geographical proximity to Polabian/Pomeranian Slavs, I'd expect admixture in the range of what see in Danes.

    Outliers aside, there isn't a significant Balto-Slavic admix in the Dutch and nearby West Germans, but they are mixed with the pre-Germanic people who lived in the area (Celts/Gauls).

    I highly doubt it's possible to find a 'pure' proto-Germanic autosomal profile nowadays. We still see guys like that during the fall of the Roman Empire/Migration period, but it looks like in the Middle-Ages Germanic speakers were already mixed with surrounding and native pre-Germanic populations.
    I know. That is a shame really. And I also agree that most Germans are a mixture of Germanic groups and continental Gaul/Celtic.

    East Germans have very visible Slavic admixture, Same thing goes with Austrians.

    Code:
    FIN_Levanluhta_IA_o:JK2065,0.135449,0.121864,0.083344,0.072352,0.047393,0.019522,-0.0094,0.005077,0.011453,0.001276,0.001137,0.002548,-0.011001,-0.013625,0.034609,0.0118,-0.004563,0.001647,0.012067,0.019384,0.000374,-0.001237,-0.004683,-0.002048,0.005389
    What about this Iron age Finn? Outliner from the Saami-like proto-Finns.

    He his quite high WHG compared to modern Germanic people.

    Target: FIN_Levanluhta_IA_o:JK2065
    Distance: 5.1637% / 0.05163707 | ADC: 0.25x RC
    48.2 Yamnaya_RUS_Samara
    28.8 TUR_Barcin_N
    23.0 WHG
    Hidden Content


    Target: Nino_scaled
    Distance: 2.3124% / 0.02312401 | R3P
    61.4 Swedish
    35.8 Spanish_Soria
    2.8 Nganassan

Page 2 of 5 FirstFirst 1234 ... LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •