Page 550 of 785 FirstFirst ... 50450500540548549550551552560600650 ... LastLast
Results 5,491 to 5,500 of 7848

Thread: Global25 automated nMonte for South/Central Asian members

  1. #5491
    Registered Users
    Posts
    3,633
    Sex
    Y-DNA (P)
    L1a1
    mtDNA (M)
    M30
    Y-DNA (M)
    R-L295

    pen=0 and using averages like you did (good suggestion) bingo 0 steppe for Midi, Reddy + South Indian non-elite

    I do suspect some level of the Tepe Hissar and Sarazm %age differences are exaggerated due to Iran_N trapping though given the high variance in geographically close pops

    Again, midi scores the highest Tepe Hissar by a significant margin

    UP Jatt scores the highest Sarazm by a significant margin

    i used samples representative of a few clusters

    Indo-Aryan elite:

    UP + Bihar + Nepal Brahmin

    post IA high steppe pops:

    Ror + UP Jatt

    NW non-elite:

    Gujjar

    West Indian:

    Gujarati + Patel + Reddy

    Gangetic non-elite:

    Gupta

    South non-elite:

    Irula
    Paniya
    North Kannadi



    Another possible trend appears: Gangetic non-elites seem to be low Sarazm!

    2 more cluster ref pops

    Gangetic non-elite

    Arlus

    West/South Indian (some Velama samples are Pirmalai shifted)

    Velama



    Loving this set up!
    Last edited by bmoney; 07-18-2019 at 03:36 AM.

  2. The Following 11 Users Say Thank You to bmoney For This Useful Post:

     26284729292 (07-18-2019),  agent_lime (07-18-2019),  Arlus (12-14-2019),  Censored (07-18-2019),  FrostAssassin0701 (07-18-2019),  Jatt1 (08-12-2019),  kush (07-18-2019),  laltota (07-18-2019),  parasar (07-18-2019),  Reza (07-18-2019),  vishankar (07-18-2019)

  3. #5492
    Banned
    Posts
    2,951
    Sex
    Location
    America
    Ethnicity
    Telgu/marathi

    Siberian Tatars Tajikistan
    Quote Originally Posted by bmoney View Post
    pen=0 and using averages like you did (good suggestion) bingo 0 steppe for Midi, Reddy + South Indian non-elite

    I do suspect some level of the Tepe Hissar and Sarazm %age differences are exaggerated due to Iran_N trapping though given the high variance in geographically close pops

    Again, midi scores the highest Tepe Hissar by a significant margin

    UP Jatt scores the highest Sarazm by a significant margin

    i used samples representative of a few clusters

    Indo-Aryan elite:

    UP + Bihar + Nepal Brahmin

    post IA high steppe pops:

    Ror + UP Jatt

    NW non-elite:

    Gujjar

    West Indian:

    Gujarati + Patel + Reddy

    Gangetic non-elite:

    Gupta

    South non-elite:

    Irula
    Paniya
    North Kannadi



    Another possible trend appears: Gangetic non-elites seem to be low Sarazm!

    2 more cluster ref pops

    Gangetic non-elite

    Arlus

    West/South Indian (some Velama samples are Pirmalai shifted)

    Velama



    Loving this set up!
    Yes that's the problem, this seems to have given more genuine steppe values for some groups at the expense of other components. Tepe Hissar is the main source of Barcin ancestry besides Sintashta and when you don't score the latter, the former is the only option. That inflates it for some users, though I dont get why the UP guy has it so low in that case.

  4. #5493
    Registered Users
    Posts
    1,052
    Ethnicity
    Brahmin (mixed)
    Nationality
    Indian
    Y-DNA (P)
    R-1A (Z-93)
    mtDNA (M)
    M-30

    Quote Originally Posted by bmoney View Post
    pen=0 and using averages like you did (good suggestion) bingo 0 steppe for Midi, Reddy + South Indian non-elite

    I do suspect some level of the Tepe Hissar and Sarazm %age differences are exaggerated due to Iran_N trapping though given the high variance in geographically close pops

    Again, midi scores the highest Tepe Hissar by a significant margin

    UP Jatt scores the highest Sarazm by a significant margin

    i used samples representative of a few clusters

    Indo-Aryan elite:

    UP + Bihar + Nepal Brahmin

    post IA high steppe pops:

    Ror + UP Jatt

    NW non-elite:

    Gujjar

    West Indian:

    Gujarati + Patel + Reddy

    Gangetic non-elite:

    Gupta

    South non-elite:

    Irula
    Paniya
    North Kannadi



    Another possible trend appears: Gangetic non-elites seem to be low Sarazm!

    2 more cluster ref pops

    Gangetic non-elite

    Arlus

    West/South Indian (some Velama samples are Pirmalai shifted)

    Velama



    Loving this set up!
    Yeah this is a very strong split. Like to the point where I think it's really significant.

  5. The Following User Says Thank You to 26284729292 For This Useful Post:

     FrostAssassin0701 (07-18-2019)

  6. #5494
    Registered Users
    Posts
    171
    Location
    NYC
    Ethnicity
    Sindhi
    Nationality
    American

    A few charts just to highlight the shifts that occurs when the penalty is set to zero (1000/500/0) and the median of the target groups is used. The Kalash end up with quite a bit less Steppe as an example.

    (warning! these are large images)

    Sorted for Sintashta
    sintashta.png
    Sorted for Tepe Hissar
    tepe.png
    Sorted for West Siberia
    wshg.png

  7. The Following 6 Users Say Thank You to heksindhi For This Useful Post:

     agent_lime (07-18-2019),  bmoney (07-22-2019),  Censored (07-18-2019),  FrostAssassin0701 (07-18-2019),  laltota (07-18-2019),  Reza (07-18-2019)

  8. #5495
    Gold Class Member
    Posts
    2,243
    Sex
    Location
    US/ India
    Ethnicity
    Punjabi Khatri/J&K/Multan
    Nationality
    India
    Y-DNA (P)
    J2b2a
    mtDNA (M)
    U2b

    India United States of America India Punjab Jammu and Kashmir
    I tried to do an elemental run. So skipping the BMAC. The AASI's are all used to pick what works best. Gradient Descent.

    [1] "distance%=2.6194"

    agentlime

    IRN_Tepe_Hissar_C,63.2
    RUS_Sidelkino_HG,12.8
    S_AASI_Sim_Hakkipikki,8.8
    Simulated_AASI_NW_by_DMXX_NW_SiSBA2,8.4
    Iberomaurusian,1.6
    WHG,1.6
    Han,1.2
    Simulated_AASI_NW_by_DMXX_NW_SGPT,1.2
    Bolivian_Cochabamba,0.8
    RUS_West_Siberia_N,0.4

    [1] "distance%=1.7134"

    Ror

    IRN_Tepe_Hissar_C,38.8
    RUS_Sidelkino_HG,21.6
    Anatolia_Barcin_N,12.4
    Simulated_AASI_NW_by_DMXX_NW_SGPT,11.6
    RUS_West_Siberia_N,8
    Simulated_AASI_NW_by_DMXX_NW_SiSBA3,6
    Simulated_AASI_NW_by_DMXX_NW_SiSBA2,1.6

    [1] "distance%=2.5271"

    Sapporo

    IRN_Tepe_Hissar_C,52.4
    RUS_Sidelkino_HG,11.2
    RUS_West_Siberia_N,10
    Simulated_AASI_NW_by_DMXX_NW_SiSBA3,8.4
    Simulated_AASI_NW_by_DMXX_NW_SGPT,6.4
    Anatolia_Barcin_N,4
    S_AASI_Sim_Hakkipikki,3.6
    Simulated_AASI_NW_by_DMXX_NW_SiSBA2,2.4
    WHG,1.2
    Levant_Canaanite_MBA,0.4

    [1] "distance%=1.837"

    poi

    IRN_Tepe_Hissar_C,25.6
    Simulated_AASI_NW_by_DMXX_NW_SGPT,15.6
    Anatolia_Barcin_N,14.8
    RUS_West_Siberia_N,13.6
    RUS_Sidelkino_HG,12.4
    S_AASI_Sim_Mala,11.6
    Han,4.4
    Simulated_AASI_NW_by_DMXX_NW_SiSBA3,2

    [1] "distance%=1.8991"

    Punjabi_Ramgarhia

    IRN_Tepe_Hissar_C,46.8
    RUS_Sidelkino_HG,14.8
    S_AASI_Sim_Hakkipikki,12
    Simulated_AASI_NW_by_DMXX_NW_SiSBA2,7.2
    Anatolia_Barcin_N,6.8
    Simulated_AASI_NW_by_DMXX_NW_SGPT,6.8
    RUS_West_Siberia_N,4.8
    Han,0.4
    NPL_Chokhopani_2700BP,0.4

    [1] "distance%=1.8431"

    Khana

    IRN_Tepe_Hissar_C,56.8
    RUS_West_Siberia_N,14.4
    Simulated_AASI_NW_by_DMXX_NW_SiSBA3,12.4
    Anatolia_Barcin_N,4
    Simulated_AASI_NW_by_DMXX_NW_SGPT,4
    WHG,3.2
    S_AASI_Sim_Mala,2.8
    Han,2
    S_AASI_Sim_Hakkipikki,0.4

    [1] "distance%=2.0772"

    Reza

    IRN_Tepe_Hissar_C,32.4
    S_AASI_Sim_Hakkipikki,18.8
    RUS_Sidelkino_HG,10
    Han,9.6
    Simulated_AASI_NW_by_DMXX_NW_SGPT,8.8
    Simulated_AASI_NW_by_DMXX_NW_SiSBA3,6.8
    RUS_West_Siberia_N,6
    S_AASI_Sim_Pulliyar,6
    Iberomaurusian,0.8
    S_AASI_Sim_Mala,0.4
    WHG,0.4

    [1] "distance%=2.3934"

    Bmoney

    IRN_Tepe_Hissar_C,48.4
    Simulated_AASI_NW_by_DMXX_NW_SiSBA3,13.6
    S_AASI_Sim_Irula,12.8
    Simulated_AASI_NW_by_DMXX_NW_SGPT,7.6
    RUS_West_Siberia_N,4.4
    Simulated_AASI_NW_by_DMXX_NW_SiSBA2,4
    RUS_Sidelkino_HG,3.6
    WHG,3.2
    Iberomaurusian,2.4

  9. The Following 3 Users Say Thank You to agent_lime For This Useful Post:

     bmoney (07-22-2019),  Reza (07-18-2019),  scobar (07-18-2019)

  10. #5496
    Banned
    Posts
    2,951
    Sex
    Location
    America
    Ethnicity
    Telgu/marathi

    Siberian Tatars Tajikistan
    Quote Originally Posted by heksindhi View Post
    A few charts just to highlight the shifts that occurs when the penalty is set to zero (1000/500/0) and the median of the target groups is used. The Kalash end up with quite a bit less Steppe as an example.

    (warning! these are large images)

    Sorted for Sintashta
    sintashta.png
    Sorted for Tepe Hissar
    tepe.png
    Sorted for West Siberia
    wshg.png
    Sarazm seems to have been omitted. I wonder if that impacted steppe or other values.

  11. #5497
    Gold Class Member
    Posts
    2,243
    Sex
    Location
    US/ India
    Ethnicity
    Punjabi Khatri/J&K/Multan
    Nationality
    India
    Y-DNA (P)
    J2b2a
    mtDNA (M)
    U2b

    India United States of America India Punjab Jammu and Kashmir
    Quote Originally Posted by Censored View Post
    Sarazm seems to have been omitted. I wonder if that impacted steppe or other values.
    Obviously it will. It's good for BMAC but is a mix of various ancestries.

    [1] "distance%=3.2501"

    TJK_Sarazm_En

    Parkhai_MBA,56
    RUS_Steppe_Maykop,23.6
    Ganj_Dareh_N,16.4
    Simulated_AASI_NW_by_DMXX,4

  12. The Following User Says Thank You to agent_lime For This Useful Post:

     Sapporo (07-18-2019)

  13. #5498
    Registered Users
    Posts
    171
    Location
    NYC
    Ethnicity
    Sindhi
    Nationality
    American

    Quote Originally Posted by Censored View Post
    Sarazm seems to have been omitted. I wonder if that impacted steppe or other values.
    Yes, that was a deliberate choice to draw out the base WSHG component, but here is a rerun with Sarazm added. Also added Gujars as they were left out in the previous run.

    Sorted for Sintashta
    sintashta2.png
    Sorted for Tepe Hissar
    tepe2.png
    Sorted for Sarazm
    sarazm.png

  14. The Following 3 Users Say Thank You to heksindhi For This Useful Post:

     Censored (07-18-2019),  laltota (07-18-2019),  Reza (07-18-2019)

  15. #5499
    Banned
    Posts
    2,951
    Sex
    Location
    America
    Ethnicity
    Telgu/marathi

    Siberian Tatars Tajikistan
    Quote Originally Posted by heksindhi View Post
    Yes, that was a deliberate choice to draw out the base WSHG component, but here is a rerun with Sarazm added. Also added Gujars as they were left out in the previous run.

    Sorted for Sintashta
    sintashta2.png
    Sorted for Tepe Hissar
    tepe2.png
    Sorted for Sarazm
    sarazm.png
    The steppe percentages look much better, although theres lots of fluctuation btw Tepe Hissar and Sarazm even within groups. I didnt intend it to be a perfect model, it's far from it. Dont think we can really have one till more aDNA is available but the goal was to finally be able to remove the noisiness of pontic steppe which I think it was able to achieve.

    Also, MA1 kind of makes WSHG redundant.
    Last edited by Censored; 07-18-2019 at 05:59 AM.

  16. The Following 2 Users Say Thank You to Censored For This Useful Post:

     FrostAssassin0701 (07-18-2019),  laltota (07-18-2019)

  17. #5500
    Gold Class Member
    Posts
    4,076
    Location
    Shangri La

    Afghanistan Jammu and Kashmir United States of America Canada
    Quote Originally Posted by agent_lime View Post
    I tried to do an elemental run. So skipping the BMAC. The AASI's are all used to pick what works best. Gradient Descent.

    [1] "distance%=2.6194"

    agentlime

    IRN_Tepe_Hissar_C,63.2
    RUS_Sidelkino_HG,12.8
    S_AASI_Sim_Hakkipikki,8.8
    Simulated_AASI_NW_by_DMXX_NW_SiSBA2,8.4
    Iberomaurusian,1.6
    WHG,1.6
    Han,1.2
    Simulated_AASI_NW_by_DMXX_NW_SGPT,1.2
    Bolivian_Cochabamba,0.8
    RUS_West_Siberia_N,0.4

    [1] "distance%=1.7134"

    Ror

    IRN_Tepe_Hissar_C,38.8
    RUS_Sidelkino_HG,21.6
    Anatolia_Barcin_N,12.4
    Simulated_AASI_NW_by_DMXX_NW_SGPT,11.6
    RUS_West_Siberia_N,8
    Simulated_AASI_NW_by_DMXX_NW_SiSBA3,6
    Simulated_AASI_NW_by_DMXX_NW_SiSBA2,1.6

    [1] "distance%=2.5271"

    Sapporo

    IRN_Tepe_Hissar_C,52.4
    RUS_Sidelkino_HG,11.2
    RUS_West_Siberia_N,10
    Simulated_AASI_NW_by_DMXX_NW_SiSBA3,8.4
    Simulated_AASI_NW_by_DMXX_NW_SGPT,6.4
    Anatolia_Barcin_N,4
    S_AASI_Sim_Hakkipikki,3.6
    Simulated_AASI_NW_by_DMXX_NW_SiSBA2,2.4
    WHG,1.2
    Levant_Canaanite_MBA,0.4

    [1] "distance%=1.837"

    poi

    IRN_Tepe_Hissar_C,25.6
    Simulated_AASI_NW_by_DMXX_NW_SGPT,15.6
    Anatolia_Barcin_N,14.8
    RUS_West_Siberia_N,13.6
    RUS_Sidelkino_HG,12.4
    S_AASI_Sim_Mala,11.6
    Han,4.4
    Simulated_AASI_NW_by_DMXX_NW_SiSBA3,2

    [1] "distance%=1.8991"

    Punjabi_Ramgarhia

    IRN_Tepe_Hissar_C,46.8
    RUS_Sidelkino_HG,14.8
    S_AASI_Sim_Hakkipikki,12
    Simulated_AASI_NW_by_DMXX_NW_SiSBA2,7.2
    Anatolia_Barcin_N,6.8
    Simulated_AASI_NW_by_DMXX_NW_SGPT,6.8
    RUS_West_Siberia_N,4.8
    Han,0.4
    NPL_Chokhopani_2700BP,0.4

    [1] "distance%=1.8431"

    Khana

    IRN_Tepe_Hissar_C,56.8
    RUS_West_Siberia_N,14.4
    Simulated_AASI_NW_by_DMXX_NW_SiSBA3,12.4
    Anatolia_Barcin_N,4
    Simulated_AASI_NW_by_DMXX_NW_SGPT,4
    WHG,3.2
    S_AASI_Sim_Mala,2.8
    Han,2
    S_AASI_Sim_Hakkipikki,0.4

    [1] "distance%=2.0772"

    Reza

    IRN_Tepe_Hissar_C,32.4
    S_AASI_Sim_Hakkipikki,18.8
    RUS_Sidelkino_HG,10
    Han,9.6
    Simulated_AASI_NW_by_DMXX_NW_SGPT,8.8
    Simulated_AASI_NW_by_DMXX_NW_SiSBA3,6.8
    RUS_West_Siberia_N,6
    S_AASI_Sim_Pulliyar,6
    Iberomaurusian,0.8
    S_AASI_Sim_Mala,0.4
    WHG,0.4

    [1] "distance%=2.3934"

    Bmoney

    IRN_Tepe_Hissar_C,48.4
    Simulated_AASI_NW_by_DMXX_NW_SiSBA3,13.6
    S_AASI_Sim_Irula,12.8
    Simulated_AASI_NW_by_DMXX_NW_SGPT,7.6
    RUS_West_Siberia_N,4.4
    Simulated_AASI_NW_by_DMXX_NW_SiSBA2,4
    RUS_Sidelkino_HG,3.6
    WHG,3.2
    Iberomaurusian,2.4
    Tepe Hissar is actually between BMAC and Seh Gabi, so its not elemental and definitely won't be the base ancestry for most SA groups primarily because of elevated ANF, in your case it does because your ratio to ANF to EHG is far higher, even ur own gradient descent showed that as you had excess 8% ANF after Sintashta, most prefer SIS1, Sarazm, Parkhai MBA, the more Steppe shifted groups take in Parkhai MBA Gonur, Jakrutan, Bustan. DMXX's observation was that most South Indians prefer SIS1 or a combo of Sarazm + SIS1, which is true because once I started throwing on more ANF rich Turan groups their fits got worse.

  18. The Following 3 Users Say Thank You to pegasus For This Useful Post:

     bmoney (07-22-2019),  DMXX (07-18-2019),  Sapporo (07-18-2019)

Page 550 of 785 FirstFirst ... 50450500540548549550551552560600650 ... LastLast

Similar Threads

  1. Replies: 67
    Last Post: 11-10-2020, 06:16 AM
  2. Replies: 87
    Last Post: 10-09-2020, 07:43 AM
  3. Automated Global25 nMonte entirely from the web
    By poi in forum Autosomal (auDNA)
    Replies: 1455
    Last Post: 08-23-2020, 07:31 AM
  4. Replies: 21
    Last Post: 07-30-2020, 12:50 PM
  5. Global25 automated nMonte for Horner members
    By drobbah in forum African
    Replies: 32
    Last Post: 06-29-2020, 12:12 PM

Posting Permissions

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