Page 1 of 2 12 LastLast
Results 1 to 10 of 17

Thread: Creating Ghost Populations (Manual + GenoPlot)

  1. #1
    Registered Users
    Posts
    25
    Sex

    Creating Ghost Populations (Manual + GenoPlot)

    How does one make ghost populations to model with?
    Last edited by DMXX; 07-29-2020 at 12:14 PM. Reason: per JRD's comment later in thread

  2. The Following User Says Thank You to JRD For This Useful Post:

     Rafe (07-27-2020)

  3. #2
    Registered Users
    Posts
    53
    Sex

    (Joking) I think they just appear and are scary.

  4. The Following 4 Users Say Thank You to RP48 For This Useful Post:

     DgidguBidgu (07-28-2020),  DMXX (07-29-2020),  Ryukendo (07-28-2020),  sheepslayer (07-27-2020)

  5. #3
    Registered Users
    Posts
    25
    Sex

    I figured it out for admixed ghost populations. It's by tweaking the coordinates. Some of the changes change only the fits, whereas others change the composition, and some change both. An example I did was making a WSHG-like population with lowered Baikal HG (though not eliminated because I haven't figured out that yet) by tweaking RUS_Tyumen_HG's coordinates, and then including them in models of the Eneolithic Piedmont Steppe. The artificial ghost ate some of the extra ANE and some of the EHG, yet the fits weren't that much higher than models not featuring it, and the distance of modeling the last draft of the ghost itself was higher than 10, so I threw it in the trash. About the joke, not that helpful but funny nevertheless.

    Edit: Tweaking the coordinates can result in unrealistically low fits or more than one components being lost. Artificial populations that have more than 10 of distance are useless for modeling. Also corrected a mix-up involving fit and distance.
    Last edited by JRD; 07-28-2020 at 03:33 PM.

  6. #4
    Registered Users
    Posts
    37

    You can make them easily on genoplot g25.

  7. #5
    Registered Users
    Posts
    25
    Sex

    Thanks for that one. Seems less time-intensive and less prone to error compared to manually tweaking the coordinates.

  8. #6
    Registered Users
    Posts
    25
    Sex

    I've successfully made a ghost population by tweaking coordinates. I named it "TUR_Agacli_HG", though it's just a very slightly tweaked (just one digit, to be specific) TUR_Pinarbasi_HG. It models as 96.6% TUR_Pinarbasi_HG, 2.6% CHG, and 0.8% Levant_Natufian. I put the scare quotes on the name because I expect actual samples from the Ağašlı culture to be different.

    Edit: I just made a better draft of the "TUR_Agacli_HG" ghost. It models as 97.6% TUR_Pinarbasi_HG and 2.4% CHG. Even then, the draft will still keep improving, so that this is the last update on it.
    Last edited by JRD; 07-29-2020 at 09:01 AM.

  9. #7
    Registered Users
    Posts
    37

    The subtraction feature on Genoplot is pretty neat too. I tried removing the Steppe and Farmer from Baltic Bronze Age to simulate the alleged Hunter Gatherers who are responsible for the so called Baltic Drift. It works much like LTU_Narva.

    Target: Latvian
    Distance: 1.4560% / 0.01455961
    47.6 Yamnaya_RUS_Samara
    30.0 Baltic_Drift_HG
    22.2 TUR_Barcin_N
    0.2 RUS_Devils_Gate_Cave_N

    Target: Russian_Smolensk
    Distance: 2.2455% / 0.02245486
    44.2 Yamnaya_RUS_Samara
    30.2 TUR_Barcin_N
    24.6 Baltic_Drift_HG
    1.0 RUS_Devils_Gate_Cave_N


    Target: Latvian
    Distance: 7.2257% / 0.07225662
    47.4 Yamnaya_RUS_Samara
    30.2 Baltic_LTU_Narva
    22.4 TUR_Barcin_N

    Target: Russian_Smolensk
    Distance: 6.4724% / 0.06472419
    45.6 Yamnaya_RUS_Samara
    30.4 TUR_Barcin_N
    24.0 Baltic_LTU_Narva

    Baltic_Drift_HG:scaled,0.146411,0.137279,0.209743, 0.195729,0.152425,0.072498,0.039552,0.060676,0.047 038,-0.07748,-0.003175,-0.053628,0.108995,0.135693,-0.07634,-0.010578,0.019469,-0.000417,-0.004626,-0.000812,0.003241,-0.020068,0.016283,-0.068312,0.013771
    Baltic_Drift_HG:unscaled.0.012863,0.013518,0.05561 7,0.060597,0.049529,0.025995,0.01683,0.026294,0.02 2999,-0.042516,-0.001955,-0.035784,0.073318,0.098598,-0.056248,-0.007978,0.014932,-0.000329,-0.00368,-0.000649,0.002597,-0.016229,0.013212,-0.056691,0.0115

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

     Coldmountains (07-29-2020)

  11. #8
    Moderator
    Posts
    1,557
    Sex
    Y-DNA (P)
    R1a-Z93
    mtDNA (M)
    H28a/H1b5(Wife)

    Quote Originally Posted by Korotyr View Post
    The subtraction feature on Genoplot is pretty neat too. I tried removing the Steppe and Farmer from Baltic Bronze Age to simulate the alleged Hunter Gatherers who are responsible for the so called Baltic Drift. It works much like LTU_Narva.

    Target: Latvian
    Distance: 1.4560% / 0.01455961
    47.6 Yamnaya_RUS_Samara
    30.0 Baltic_Drift_HG
    22.2 TUR_Barcin_N
    0.2 RUS_Devils_Gate_Cave_N

    Target: Russian_Smolensk
    Distance: 2.2455% / 0.02245486
    44.2 Yamnaya_RUS_Samara
    30.2 TUR_Barcin_N
    24.6 Baltic_Drift_HG
    1.0 RUS_Devils_Gate_Cave_N


    Target: Latvian
    Distance: 7.2257% / 0.07225662
    47.4 Yamnaya_RUS_Samara
    30.2 Baltic_LTU_Narva
    22.4 TUR_Barcin_N

    Target: Russian_Smolensk
    Distance: 6.4724% / 0.06472419
    45.6 Yamnaya_RUS_Samara
    30.4 TUR_Barcin_N
    24.0 Baltic_LTU_Narva

    Baltic_Drift_HG:scaled,0.146411,0.137279,0.209743, 0.195729,0.152425,0.072498,0.039552,0.060676,0.047 038,-0.07748,-0.003175,-0.053628,0.108995,0.135693,-0.07634,-0.010578,0.019469,-0.000417,-0.004626,-0.000812,0.003241,-0.020068,0.016283,-0.068312,0.013771
    Baltic_Drift_HG:unscaled.0.012863,0.013518,0.05561 7,0.060597,0.049529,0.025995,0.01683,0.026294,0.02 2999,-0.042516,-0.001955,-0.035784,0.073318,0.098598,-0.056248,-0.007978,0.014932,-0.000329,-0.00368,-0.000649,0.002597,-0.016229,0.013212,-0.056691,0.0115
    Balto-Slavic drift is not just caused by HG ancestry. It shows here up as HG because you removed Steppe and Farmer. But this component will include very western GAC WHG, SHG/WHG-like Baltic_HG and more eastern Ukraine_N from Corded Ware, what in the end is again similar to Baltic_HG. Still it seems to be here mainly WHG-like.

    Around 10% of WHG ancestry of Balto-Slavs should be from GAC based on them being around 30-40% GAC-derived. So the extra HG should rather be subtracted from Early CWC and GAC but Balto-Slavic drift will not make the extract very close to any existing population.

  12. #9
    Registered Users
    Posts
    37

    Quote Originally Posted by Coldmountains View Post
    Balto-Slavic drift is not just caused by HG ancestry. It shows here up as HG because you removed Steppe and Farmer. But this component will include very western GAC WHG, SHG/WHG-like Baltic_HG and more eastern Ukraine_N from Corded Ware, what in the end is again similar to Baltic_HG. Still it seems to be here mainly WHG-like.
    Can you expand on that? Ultimately, is the drift caused by accumulated HG ancestries from different sources?
    Last edited by Korotyr; 07-29-2020 at 12:01 PM.

  13. #10
    Registered Users
    Posts
    25
    Sex

    Now that I figured out how to make them by both tweaking coordinates and Genoplot, we could turn this thread into a more general ghost populations one, though the title would become non-indicative.

Page 1 of 2 12 LastLast

Similar Threads

  1. Ghost feature experimentation
    By BalkanKiwi in forum GenoPlot
    Replies: 0
    Last Post: 07-05-2020, 09:35 AM
  2. 'Ghost' DNA discovered in West Africans
    By Elizabeth in forum Ancient (aDNA)
    Replies: 15
    Last Post: 02-18-2020, 06:29 PM
  3. Replies: 17
    Last Post: 07-06-2018, 09:02 PM
  4. In saliva, clues to a 'ghost' species of ancient human
    By Amerijoe in forum Human Evolution
    Replies: 1
    Last Post: 07-22-2017, 02:28 AM
  5. Ghost Hunters
    By Kale in forum Ancient (aDNA)
    Replies: 3
    Last Post: 08-01-2016, 03:52 AM

Posting Permissions

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