Page 15 of 18 FirstFirst ... 51314151617 ... LastLast
Results 141 to 150 of 171

Thread: New Simulated AASI G25 Coordinates (+ Updates)

  1. #141
    Bronze Class Member
    Posts
    2,188
    Sex
    Location
    Krasnoyarsk

    Siberian Tatars Tajikistan
    What would a typical Kushan genome(predictably) have looked like? As far as Siberian HG, Sintashta steppe, BMAC, etc?

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

     26284729292 (07-22-2019)

  3. #142
    Registered Users
    Posts
    183
    Sex
    Location
    UK
    Ethnicity
    Punjabi (Jatt)
    Y-DNA
    R2b (R-Y136220)
    mtDNA
    M3a1

    Quote Originally Posted by Censored View Post
    For Asians the unscaled seems to provide more reasonable results more in line with what other sources say. As far as giving more “exotic results” I have seen no penalty have the opposite effect in many cases, ie removing them.
    Which other sources?

  4. #143
    Registered Users
    Posts
    289
    Sex
    Ethnicity
    Malayali Muslim
    Nationality
    Indian
    Y-DNA
    HM82->Z4507->Z5890
    mtDNA
    R30a

    Quote Originally Posted by Huijbregts View Post
    I recommend not to use the penalty option unless you have an explicit reason to do so and completely understand what you are doing.
    For instance, when Davidski has estimated a qpAdm model and he wants to compare it with an nMonte model, I can imagine that he prefers to drop the penalizing.

    I have included the penalizing in nMonte3, because I noted that nMonte is vulnerable to overfitting. As a consequence of the ever growing content of the Global25, this vulnerability will only grow (but many thanks, David).
    nMonte is a simple algorithm (which often is an advantage!) and so is the penalizing:
    The nMonte algorithm follows a random walk which gradually improves the fit of the mixture. The random walk is guided by minimizing a misfit formula; this formula is not applied to the individual samples, but to the average of the present mixture.
    In nMonte3 I have supplemented this misfit function with a small extra term which penalizes distant individual samples. This should largely prevent the inclusion of oddball overfitting; it appears to be quite effective.

    I too am surprised about the ineradicable persistence of pen=0. Presumably some people think that they are using an advanced feature by declaring pen=0; actually they are disabling a useful feature of nMonte3.
    But I think the main reason is that many people love the overfitted oddball admixtures from pen=0. It is exciting to have Scythian or Amerindian admixture.

    By the way, I am not a mathematician but a retired biologist. I got a university training in 20-century statistics and some modelling experience in my professional career. Most of my present knowledge about data modelling is from internet sources; obviously I am not an expert.
    Can pen=0 be more useful for mixed individuals or when modeling w/ ancients because it doesn't penalize distant admixtures? You mentioned Davidski and I believe he uses pen=0 on the scaled ancients model he hands out with your G25 co-ords.

    Had another question about the Nbatch parameter too that would be really helpful..I noticed w/ Davidski's scaled ancients model (actually a variation of it using Dmx's aasi sims) that using an nbatch=200 and with the default penalty seemed to consistently produce a non-trivial better fit instead of the default 500 nbatch in nmonte3. The model is just 10 pops or so..I don't know if this is an artifact of the particular model, samples being modeled, or just the size of the model and params being used or something else? Higher vs lower nbatch didn't affect the pen=0 runs much fyi.
    Last edited by scobar; 07-22-2019 at 08:25 AM.

  5. #144
    Registered Users
    Posts
    943
    Sex
    Location
    US
    Ethnicity
    India Andhra Kamma Telugu
    Nationality
    US
    Y-DNA
    HM82 Z5888+CTS8144+
    mtDNA
    U2a1a

    United States of America India
    Quote Originally Posted by scobar View Post
    Can pen=0 be more useful for mixed individuals or when modeling w/ ancients because it doesn't penalize distant admixtures? You mentioned Davidski and I believe he uses pen=0 on the scaled ancients model he hands out with your G25 co-ords.

    Had another question about the Nbatch parameter too that would be really helpful..I noticed w/ Davidski's scaled ancients model (actually a variation of it using Dmx's aasi sims) that using an nbatch=200 and with the default penalty seemed to consistently produce a non-trivial better fit instead of the default 500 nbatch in nmonte3. The model is just 10 pops or so..I don't know if this is an artifact of the particular model, samples being modeled, or just the size of the model and params being used or something else? Higher vs lower nbatch didn't affect the pen=0 runs much fyi.
    zero penalty & lesser nbatch might produce better for Europeans but not for Indians since mix is very complex with multiple pulls. Compare whata you get with big penalty close to default & big batch you will see

  6. The Following 2 Users Say Thank You to tipirneni For This Useful Post:

     pegasus (07-23-2019),  scobar (07-23-2019)

  7. #145
    Registered Users
    Posts
    289
    Sex
    Ethnicity
    Malayali Muslim
    Nationality
    Indian
    Y-DNA
    HM82->Z4507->Z5890
    mtDNA
    R30a

    Tried modifying a couple of existing models using the new AASI sims avgs - Davidski's ancients model & an eneolithic model posted by Sorcelow. So swapped in the sims & West_Siberia_N for Sisba1/2/3 in the former and just the sims for Sisba3 in the latter. Here they are & some runs with them, including runs of some ancients against the models too.

    Models

     
    Sorcelow's N/EN model -

    Code:
    ,PC1,PC2,PC3,PC4,PC5,PC6,PC7,PC8,PC9,PC10,PC11,PC12,PC13,PC14,PC15,PC16,PC17,PC18,PC19,PC20,PC21,PC22,PC23,PC24,PC25
    STEPPE-Eneolithic:Vonyuchka_Eneolithic:VJ1001,0.110408,0.081242,0.006034,0.098838,-0.063704,0.045738,0.007755,-0.014076,-0.077105,-0.082735,0.003248,0.002398,-0.001784,-0.016928,0.037323,0.01432,-0.016298,-0.000887,-0.001131,0.003627,0.009858,-0.000618,0.006779,-0.005663,-0.008502
    STEPPE-Eneolithic:Progress_Eneolithic:PG2001,0.112685,0.079211,0.010559,0.106268,-0.051702,0.043507,0.000705,-0.007846,-0.07731,-0.077997,0.004384,0.003447,-0.008028,-0.022845,0.033794,0.005436,-0.009909,-0.005954,-0.006034,0.006503,-0.007736,0.003833,0.00419,0.005422,-0.007784
    STEPPE-Eneolithic:Progress_Eneolithic:PG2004,0.114961,0.071087,0.023381,0.127263,-0.051086,0.056615,-0.003995,-0.011769,-0.066675,-0.089113,0.008444,0.006744,-0.004162,-0.020643,0.03203,0.004641,-0.010561,-0.000887,-0.003268,0.005878,-0.003494,0.007172,0.00986,0.008314,0.002754
    HUNTERS-CHG:KK1,0.087644,0.106631,-0.086361,-0.002261,-0.08617,0.0251,0.028436,0.003923,-0.129464,-0.077086,-0.005359,0.023979,-0.056788,0.012386,0.030401,-0.032617,0.025034,-0.012289,-0.021117,0.033641,0.030821,-0.009398,0.009613,-0.022654,-0.001078
    HUNTERS-WHG:Loschbour:Loschbour,0.130897,0.109677,0.203268,0.198323,0.1628,0.059404,0.014571,0.038537,0.100217,0.016401,-0.015589,-0.017684,0.019772,-0.000826,0.061346,0.070272,0.002738,0.007475,-0.009302,0.065656,0.118042,0.010881,-0.048929,-0.174723,0.019399
    HUNTERS-WHG:I1875,0.130897,0.120848,0.191577,0.194447,0.159107,0.048248,0.015746,0.040152,0.087332,0.007472,-0.016076,-0.015436,0.017839,-0.003303,0.047909,0.047069,0.007302,0.018497,-0.003268,0.057027,0.087471,0.00915,-0.048436,-0.143153,0.01449
    HUNTERS-West_Siberia_Neolithic:I1960,0.103579,-0.059916,0.100314,0.202199,-0.091094,0.044901,-0.048882,-0.054921,-0.035178,-0.090571,0.028093,-0.011989,0.02661,-0.077619,0.028637,0.01485,-0.015646,-0.004814,-0.004399,0.001376,-0.025705,0.011747,0.02465,0.00976,-0.013053
    HUNTERS-West_Siberia_Neolithic:I5766,0.108132,-0.060932,0.094657,0.196385,-0.074168,0.046575,-0.035017,-0.03946,-0.031701,-0.093669,0.01088,-0.009292,0.028097,-0.066197,0.023208,0.01843,-0.011474,-0.008361,0.003645,0.005127,-0.034564,0.01014,0.02354,0.005181,-0.007185
    PASTORALISTS-Iran_Neolithic:Wezmeh_Cave_N:WC1,0.037562,0.072103,-0.165556,-0.016473,-0.11756,0.01506,0.017861,-0.003231,-0.071583,-0.046835,0.003085,-0.003147,0.00669,0.000688,0.022122,0.053036,-0.011735,0.001014,0.014581,-0.035267,0.009358,-0.023741,-0.006655,-0.032896,0.019998
    PASTORALISTS-Iran_Neolithic:Abdul_Hosein_N:AH2,0.03187,0.064994,-0.160653,-0.00969,-0.115406,0.011713,0.010105,-0.005538,-0.079969,-0.051755,-0.007795,-0.005545,0.004608,0.002615,0.032709,0.055422,-0.010691,0.011022,0.014707,-0.03189,0.004492,-0.025102,-0.010969,-0.044223,0.020956
    LEVANT-Neolithic:I1170,0.087644,0.165531,-0.047517,-0.141475,0.00954,-0.066097,-0.014806,-0.013615,0.047859,0.035354,0.014777,-0.012589,0.039841,0.007982,-0.02443,0.003978,0.001956,0.004687,0.004274,0.011756,-0.006863,0.013973,-0.003574,-0.000241,-0.002395
    IBEROMAURUSIAN:TAF013,-0.196914,0.084289,-0.027153,-0.088502,0.032314,-0.057173,-0.072148,0.015461,0.157279,0.000911,0.017538,-0.032971,0.079831,-0.052709,0.066367,-0.041766,0.002347,-0.065118,-0.141536,0.035642,-0.037559,-0.125013,0.078386,-0.011809,0.021435
    IBEROMAURUSIAN:TAF014,-0.196914,0.083273,-0.021873,-0.081719,0.031083,-0.054941,-0.074498,0.013846,0.162392,0.002551,0.020299,-0.03327,0.079087,-0.05092,0.070032,-0.021214,0.016689,-0.066512,-0.14669,0.04202,-0.04255,-0.119449,0.073702,-0.018195,0.014969
    FARMERS-Balkans_Neolithic:I3498,0.119514,0.188888,-0.003394,-0.109498,0.053548,-0.054384,-0.01034,-0.004384,0.045404,0.088931,0.001461,0.01124,-0.017096,-0.007707,-0.051031,-0.014585,0.009518,-0.00114,0.01081,-0.019384,-0.015597,0.006183,0.000493,0.007471,-0.001796
    FARMERS-Balkans_Neolithic:I2533,0.126344,0.187873,0.010936,-0.102068,0.060319,-0.055778,0.003055,-0.005769,0.044586,0.095492,0.001299,0.012589,-0.027948,-0.001376,-0.050488,-0.006762,0.034421,-0.000507,0.008547,-0.013381,-0.014974,0.007048,-0.003944,0.005904,-0.011137
    FARMERS-Balkans_Neolithic:I3947,0.121791,0.192951,0.010182,-0.097869,0.06155,-0.048248,-0.00611,-0.004154,0.055221,0.092029,0.008119,0.018134,-0.027056,0.005505,-0.044245,-0.023336,0.00691,0.006968,0.013198,-0.01063,-0.003619,0.008656,-0.006902,-0.014098,-0.010418
    FARMERS-Barcin_Neolithic:I1097,0.120652,0.18178,-0.001886,-0.106591,0.054472,-0.048248,-0.00611,-0.007615,0.038246,0.084922,0.008607,0.014837,-0.022894,0.000413,-0.048859,-0.006629,0.021383,0.003294,0.013827,-0.015257,-0.013975,0.005564,-0.011462,-0.009278,-0.002515
    FARMERS-Barcin_Neolithic:I1581,0.114961,0.179749,0,-0.111759,0.060319,-0.048806,-0.00141,-0.00923,0.043359,0.085833,0.005521,0.015886,-0.021853,-0.003028,-0.043838,-0.000133,0.040158,0.003547,0.008799,-0.003752,-0.010482,0.00779,-0.011339,0.003615,-0.003592
    EAST-ASIATIC_Neolithic:Lokomotiv_N:DA357,0.03187,-0.415352,0.086361,-0.028747,-0.082785,-0.04769,0.002115,0.007154,0.008795,0.016037,-0.011205,-0.008093,-0.008771,-0.004954,-0.009636,-0.006629,-0.006128,0.002914,-0.001131,0.024262,-0.015722,-0.013231,-0.023787,-0.008435,0.011975
    EAST-ASIATIC_Neolithic:Shamanka_N:DA245,0.022765,-0.420429,0.090132,-0.02907,-0.089555,-0.047411,0.006345,0.011769,0.015135,0.016583,-0.019974,0.002398,0.006392,-0.007707,-0.007465,-0.003845,0.004955,0,0.003771,0.021385,-0.01984,-0.002102,-0.019103,-0.003494,0.005508
    SOUTH-AMERICA:Brazil_LapaDoSanto_9600BP:CP18,0.052359,-0.298566,0.118039,0.097223,-0.116637,-0.011992,-0.278723,-0.32191,-0.016771,-0.018588,0.005359,-0.004346,0.001933,0.018992,0.000407,0.008618,0.00013,-0.007728,0.003645,-0.001751,0,0,0.011216,-0.006748,-0.001796
    SOUTH-AMERICA:Chile_LosRieles_10900BP:I11974,0.044391,-0.304659,0.118416,0.101099,-0.111098,-0.012829,-0.291648,-0.349832,-0.010022,-0.022597,0.005359,-0.003897,-0.004014,0.018854,-0.008822,0.006895,0.008475,0.000507,0.000754,0.008004,0.001248,-0.000989,0.005916,-0.000602,-0.000958
    NORTH-AMERICA:Kennewick:kennewick,0.046667,-0.307705,0.109742,0.08721,-0.100634,-0.019801,-0.220675,-0.261681,-0.01268,-0.017677,0.001949,0.000599,0.001784,0.012524,-0.010043,0.010209,0.007302,0.005701,0.013575,0.005002,0.000499,-0.008532,-0.002342,0.001566,-0.001676
    AFRICA:Malawi_Chencherere_5200BP:I4421,-0.563424,0.054839,-0.000754,0.01938,-0.004308,-0.015618,0.149467,-0.110765,0.024952,-0.021322,-0.001299,-0.031022,-0.034935,-0.008945,0.019272,-0.023203,0.017863,0.129856,-0.059707,-0.006003,-0.003743,-0.024483,0.00037,-0.00494,-0.007664
    PACIFIC:Lapita_Vanuatu:I1370,0.009106,-0.423476,-0.041483,-0.063308,0.13818,0.066376,-0.00047,-0.010153,-0.007976,-0.030798,0.060084,0.020082,-0.003717,-0.006606,0.014794,0.005038,-0.002477,0.012922,0.010433,-0.028138,0.005241,-0.018177,-0.002095,-0.011086,-0.070533
    AUSTRALIAN:B_Australian-3,-0.043253,-0.233572,-0.211942,0.232238,0.140026,-0.342478,-0.00282,0.006923,-0.006749,-0.001822,-0.005846,-0.001349,-0.003271,-0.001239,-0.002443,0.000265,0.010431,-0.002787,-0.003771,0.002501,0.007112,0.002349,-0.002095,-0.001687,-0.009101
    AUSTRALIAN:IHW9193,-0.0387,-0.22951,-0.203645,0.22707,0.143104,-0.333832,0.00611,0.003692,-0.009204,0.002005,-0.007957,-3e-04,0.003865,-0.000275,-0.005429,-0.005171,0.008736,0.001774,-0.00088,0.009004,0.005241,0.008779,0.002958,0.003133,0.001796
    S_AASI_Sim_Avg,-0.007017607,-0.247268894,-0.229976338,0.16563825,-0.035368871,0.064331863,-0.015188881,0.027727614,0.147380455,0.101064071,-0.004664075,-0.004333522,0.003228897,0.026031903,-0.047720864,-0.045200255,0.010093749,0.000906224,0.000245098,0.039040136,0.009429353,0.024010206,-0.003068505,0.015983546,-0.010868946
    NW_AASI_Sim_Avg,0.058062278,-0.24745847,-0.325699695,0.273535685,-0.119223593,0.177284363,0.005596483,0.058361388,0.170617118,0.13535108,-0.00361402,0.035724435,-0.030872625,0.03748403,-0.013375715,-0.129868563,0.005831905,-0.02091509,-0.021039598,0.054171475,0.001207383,0.0099081,0.01941486,0.05655002,-0.064804895
    Davidski's ancients model -

    Code:
    ,PC1,PC2,PC3,PC4,PC5,PC6,PC7,PC8,PC9,PC10,PC11,PC12,PC13,PC14,PC15,PC16,PC17,PC18,PC19,PC20,PC21,PC22,PC23,PC24,PC25
    Armenia_EBA,0.107753,0.128634,-0.065242,-0.045113,-0.048932,-0.00344,0.009479,-0.005154,-0.056244,-0.016948,0.003139,0.009142,-0.014222,0.003165,0.005338,-0.000442,0.012256,-0.000971,-0.002598,0.000292,0.007528,-0.004122,0.004642,-0.00237,-0.001597
    Barcin_N,0.117977651830018,0.181018118861759,0.00271526882647004,-0.101018734131967,0.051948182859461,-0.0472580222655371,-0.00442995052737612,-0.0075689153780446,0.0367836121278892,0.0809766157370447,0.00854163106203961,0.0121616751868318,-0.0229383440771125,0.000770686966283977,-0.0411978972527725,-0.00909564603532921,0.0231562069432798,0.000741128278370216,0.0121550015220073,-0.00919816194138807,-0.0131143576167497,0.00612699324505911,-0.00388230311799581,-0.00332577136917137,-0.00513726380478947
    Clovis:Anzick,0.048944,-0.294504,0.114645,0.0969,-0.097864,-0.012271,-0.259922,-0.314756,-0.011862,-0.024055,0.006496,-0.002398,-0.000743,0.010597,-0.012079,0.006232,0.008345,0.00038,-0.003268,0.005753,-0.001747,0.003586,-0.006532,-0.000723,0.003712
    Dinka,-0.577462,0.051115,0.000754,-0.007591,-0.005899,-0.001255,-0.016647,0.019691,0.080787,-0.097466,-0.019865,0.022055,-0.039569,-0.001216,0.009478,-0.021391,0.017059,-0.011212,0.022081,-0.022386,0.001996,0.003978,0.000924,0.001366,0.009799
    Ethiopia_4500BP,-0.511066,0.043668,0.000754,0.000969,-0.00277,-0.011435,0.050997,-0.045229,0.089172,-0.087838,-0.012991,-0.002997,-0.031219,0.000688,0.02158,-0.029965,0.027772,0.039273,0.00176,-0.009004,0.000374,0.006183,-0.003451,-0.00241,-0.000838
    Ganj_Dareh_N,0.044012,0.06804,-0.15374,0.002369,-0.121869,0.019057,0.017391,-0.003307,-0.083037,-0.0554,-0.000325,-0.004796,0.006145,-0.009496,0.029587,0.058826,-0.008084,0.009882,0.011103,-0.036226,0.0094,-0.02708,-0.011626,-0.037917,0.025746
    Han,0.0209434519408812,-0.450691955965491,-0.00871148748492472,-0.0652140125379508,0.0789686239439437,0.038263798494138,0.00256161595482227,-0.00553823076442288,-0.0150734068047008,-0.00328024999047329,-0.0470601650528342,-0.00666906402728299,0.00726950761743875,-0.00761053379205427,-0.0042209013492381,0.00070272483946421,0.00132991729066134,-0.00111485963242015,-0.00177234251768669,-0.0103674728068127,0.011829125614347,0.00704820615476023,0.0146541536739588,-0.000771193360967274,-0.00246684462421126
    Iberomaurusian,-0.195093,0.083883,-0.022024,-0.086048,0.030036,-0.057284,-0.074357,0.017953,0.157811,0.003608,0.018577,-0.032521,0.078196,-0.05037,0.071036,-0.039459,0.005268,-0.066765,-0.142692,0.034592,-0.03958,-0.123133,0.071188,-0.013303,0.017939
    Levant_N,0.00615,0.0168,-0.0071,-0.04625,0.01025,-0.02515,-0.00645,-0.00645,0.03675,0.02375,0.0052,-0.0088,0.02085,-0.0031,-0.01465,0.00595,0.00775,-0.00645,-0.00725,0.01445,-0.0042,0.0059,0.00125,-9e-04,-0.0037
    Nganassan,0.049399,-0.406009,0.154242,0.001809,-0.158922,-0.085731,0.027966,0.043014,0.03031,0.012429,0.103409,0.009532,-0.003152,-0.027112,-0.022095,-0.013842,-0.000991,0.013733,0.02426,-0.001476,0.043548,-0.012736,0.034559,0.002362,0.013268
    Tepecik_Ciftlik_N,0.0097,0.01705,-0.00375,-0.0302,0.00815,-0.01225,7e-04,-0.003,0.01125,0.0318,0.0094,0.0076,-0.00785,0.00045,-0.0276,0.00355,0.01325,-0.00045,0.009,4e-04,-8e-04,0.00415,-0.0067,-0.00395,-0.00655
    WHG,0.124636303669918,0.116278118427329,0.1847891284681,0.189278907118569,0.154644323974403,0.0464352948197812,0.013160595731197,0.0372676778522623,0.0890701311186865,0.0177680207817303,-0.0153457059954894,-0.0158109270759181,0.0159066935596308,-0.00302769879611562,0.0533380781431052,0.0582068310424129,0.00501978585200604,0.0163428287025227,-0.00930165576658264,0.0555891636022705,0.094458313186294,0.0111905729299263,-0.0496072065077243,-0.160866115139267,0.0170044629435922
    Yoruba,-0.630062125250819,0.0625007580664627,0.0221128476262357,0.0167079010725635,0.000503478833874871,0.0124739425309571,-0.0444170105927898,0.0477672403431473,-0.0488811971007258,0.032769332932607,0.00462076258308626,0.000790246620581196,0.0230559346809037,0.000950835044652857,0.012523319298979,-0.00960677891384516,0.00707633344275127,0.000449111067832891,0.00602206792055686,-0.00299005919587556,0.00155413248445877,0.00231564485368763,-0.00175911467946521,-0.000471030445003293,-0.000424512824894608
    Yamnaya_RUS_Samara,0.1259169,0.0887318,0.0428032,0.1164425,-0.028236,0.045529,0.0036719,-0.002279,-0.0558095,-0.0715732,0.0009742,-0.0003934,-0.0019885,-0.025529,0.0362032,0.0160764,-0.0005869,-0.0012038,-0.0038966,0.0149446,-0.0022928,6.19e-05,0.0112925,0.0189332,-0.0039516
    S_AASI_Sim_Avg,-0.007017607,-0.247268894,-0.229976338,0.16563825,-0.035368871,0.064331863,-0.015188881,0.027727614,0.147380455,0.101064071,-0.004664075,-0.004333522,0.003228897,0.026031903,-0.047720864,-0.045200255,0.010093749,0.000906224,0.000245098,0.039040136,0.009429353,0.024010206,-0.003068505,0.015983546,-0.010868946
    NW_AASI_Sim_Avg,0.058062278,-0.24745847,-0.325699695,0.273535685,-0.119223593,0.177284363,0.005596483,0.058361388,0.170617118,0.13535108,-0.00361402,0.035724435,-0.030872625,0.03748403,-0.013375715,-0.129868563,0.005831905,-0.02091509,-0.021039598,0.054171475,0.001207383,0.0099081,0.01941486,0.05655002,-0.064804895
    West_Siberia_N,0.1058555,-0.060424,0.0974855,0.199292,-0.082631,0.045738,-0.0419495,-0.0471905,-0.0334395,-0.09212,0.0194865,-0.0106405,0.0273535,-0.071908,0.0259225,0.01664,-0.01356,-0.0065875,-0.000377,0.0032515,-0.0301345,0.0109435,0.024095,0.0074705,-0.010119


    Some ancients runs using pen=0.001

     


    "distance%=2.9873" - IRN_Shahr_I_Sokhta_BA1

    Ganj_Dareh_N,74.6
    Armenia_EBA,14.4
    Yamnaya_RUS_Samara,4.8
    NW_AASI_Sim_Avg,2.6
    West_Siberia_N,1.6
    S_AASI_Sim_Avg,1.2
    Barcin_N,0.6
    Clovis,0.2

    "distance%=3.6427" - IRN_Shahr_I_Sokhta_BA1

    PASTORALISTS-Iran_Neolithic,74.4
    STEPPE-Eneolithic,12.6
    HUNTERS-CHG,7.6
    NW_AASI_Sim_Avg,1.6
    FARMERS-Balkans_Neolithic,1.2
    S_AASI_Sim_Avg,1
    FARMERS-Barcin_Neolithic,0.4
    HUNTERS-West_Siberia_Neolithic,0.4
    NORTH-AMERICA,0.4
    EAST-ASIATIC_Neolithic,0.2
    SOUTH-AMERICA,0.2


    "distance%=2.7039" - IRN_Shahr_I_Sokhta_BA3

    Ganj_Dareh_N,45.4
    S_AASI_Sim_Avg,26
    NW_AASI_Sim_Avg,20
    West_Siberia_N,3
    Tepecik_Ciftlik_N,2
    Yamnaya_RUS_Samara,2
    Clovis,0.4
    WHG,0.4
    Yoruba,0.4
    Armenia_EBA,0.2
    Nganassan,0.2

    "distance%=3.1463" - IRN_Shahr_I_Sokhta_BA3

    PASTORALISTS-Iran_Neolithic,43.2
    S_AASI_Sim_Avg,26.2
    NW_AASI_Sim_Avg,19.6
    STEPPE-Eneolithic,6
    HUNTERS-West_Siberia_Neolithic,2.8
    NORTH-AMERICA,0.8
    SOUTH-AMERICA,0.6
    HUNTERS-CHG,0.4
    AFRICA,0.2
    HUNTERS-WHG,0.2


    "distance%=1.9568" - IRN_Tepe_Hissar_C

    Ganj_Dareh_N,64
    Armenia_EBA,32
    Barcin_N,2.6
    Yamnaya_RUS_Samara,1.4

    "distance%=2.636" - IRN_Tepe_Hissar_C

    PASTORALISTS-Iran_Neolithic,65.6
    HUNTERS-CHG,14.2
    STEPPE-Eneolithic,9
    LEVANT-Neolithic,4.6
    FARMERS-Balkans_Neolithic,4.2
    FARMERS-Barcin_Neolithic,2.4





    Some pop avgs using pen=0.001

     


    Brahmin_Uttar_Pradesh
     
    "distance%=3.8803" - Brahmin_Uttar_Pradesh

    Ganj_Dareh_N,34.6
    S_AASI_Sim_Avg,17.8
    Yamnaya_RUS_Samara,14.6
    NW_AASI_Sim_Avg,13.6
    West_Siberia_N,7.8
    Armenia_EBA,4.2
    Tepecik_Ciftlik_N,3
    Barcin_N,1.6
    Levant_N,1.6
    WHG,1.2

    "distance%=3.9265" - Brahmin_Uttar_Pradesh

    PASTORALISTS-Iran_Neolithic,31.4
    STEPPE-Eneolithic,25.4
    S_AASI_Sim_Avg,17.6
    NW_AASI_Sim_Avg,13
    HUNTERS-West_Siberia_Neolithic,6.6
    FARMERS-Balkans_Neolithic,1.8
    FARMERS-Barcin_Neolithic,1.8
    HUNTERS-WHG,1.2
    AUSTRALIAN,0.4
    IBEROMAURUSIAN,0.4
    PACIFIC,0.4


    Kalash
     
    "distance%=3.5864" - Kalash

    Ganj_Dareh_N,41.2
    Yamnaya_RUS_Samara,22.6
    Armenia_EBA,10.2
    West_Siberia_N,8.2
    NW_AASI_Sim_Avg,4.8
    S_AASI_Sim_Avg,3.4
    Tepecik_Ciftlik_N,3.4
    Levant_N,1.8
    Barcin_N,1.4
    WHG,1
    Han,0.8
    Nganassan,0.8
    Clovis,0.4

    "distance%=3.8363" - Kalash

    STEPPE-Eneolithic,39.2
    PASTORALISTS-Iran_Neolithic,37
    HUNTERS-West_Siberia_Neolithic,6.4
    NW_AASI_Sim_Avg,4.4
    FARMERS-Balkans_Neolithic,3.6
    S_AASI_Sim_Avg,2.4
    FARMERS-Barcin_Neolithic,1.8
    LEVANT-Neolithic,1.6
    AUSTRALIAN,1.2
    EAST-ASIATIC_Neolithic,1.2
    NORTH-AMERICA,0.4
    PACIFIC,0.4
    HUNTERS-WHG,0.2
    SOUTH-AMERICA,0.2


    Maratha
     

    "distance%=2.844" - Maratha

    S_AASI_Sim_Avg,33.6
    Ganj_Dareh_N,21.8
    NW_AASI_Sim_Avg,17.8
    Tepecik_Ciftlik_N,6.4
    Yamnaya_RUS_Samara,6.2
    Levant_N,4.2
    Armenia_EBA,3.8
    West_Siberia_N,3.2
    Barcin_N,1.6
    Ethiopia_4500BP,0.6
    Clovis,0.4
    WHG,0.4

    "distance%=2.53" - Maratha

    S_AASI_Sim_Avg,38.4
    PASTORALISTS-Iran_Neolithic,24.8
    NW_AASI_Sim_Avg,14.6
    STEPPE-Eneolithic,7.8
    HUNTERS-West_Siberia_Neolithic,5
    FARMERS-Balkans_Neolithic,2.4
    HUNTERS-CHG,1.8
    FARMERS-Barcin_Neolithic,1.2
    LEVANT-Neolithic,1.2
    HUNTERS-WHG,1
    IBEROMAURUSIAN,1
    AFRICA,0.6
    SOUTH-AMERICA,0.2


    Kshatriya
     

    "distance%=3.0845" - Kshatriya

    Ganj_Dareh_N,35.4
    S_AASI_Sim_Avg,18.6
    Tepecik_Ciftlik_N,14.4
    NW_AASI_Sim_Avg,13
    Yamnaya_RUS_Samara,9.4
    West_Siberia_N,7.2
    Armenia_EBA,0.8
    Barcin_N,0.6
    WHG,0.6

    "distance%=3.3339" - Kshatriya

    PASTORALISTS-Iran_Neolithic,33.4
    STEPPE-Eneolithic,21.4
    S_AASI_Sim_Avg,19.8
    NW_AASI_Sim_Avg,11.8
    HUNTERS-West_Siberia_Neolithic,5.2
    FARMERS-Barcin_Neolithic,2
    FARMERS-Balkans_Neolithic,1.2
    HUNTERS-WHG,1
    AUSTRALIAN,0.8
    EAST-ASIATIC_Neolithic,0.8
    LEVANT-Neolithic,0.8
    PACIFIC,0.8
    IBEROMAURUSIAN,0.6
    NORTH-AMERICA,0.2
    SOUTH-AMERICA,0.2



    Madiga
     


    "distance%=4.8834" - Madiga, without swapping in aasi sims & west_siberia_n in Davidski's model

    Shahr_I_Sokhta_BA3,81.6
    Jarawa,18.4

    "distance%=4.3151" - Madiga

    S_AASI_Sim_Avg,44
    NW_AASI_Sim_Avg,16.4
    Ganj_Dareh_N,13.6
    Armenia_EBA,6
    Tepecik_Ciftlik_N,4.8
    Levant_N,3.8
    Yamnaya_RUS_Samara,3.6
    West_Siberia_N,3.4
    Barcin_N,1.4
    Clovis,0.8
    Iberomaurusian,0.8
    Dinka,0.4
    WHG,0.4
    Yoruba,0.4
    Ethiopia_4500BP,0.2

    "distance%=3.7267" - Madiga

    S_AASI_Sim_Avg,48.4
    PASTORALISTS-Iran_Neolithic,16.6
    NW_AASI_Sim_Avg,12.8
    STEPPE-Eneolithic,7.6
    HUNTERS-West_Siberia_Neolithic,3.8
    HUNTERS-CHG,3.4
    FARMERS-Balkans_Neolithic,1.8
    IBEROMAURUSIAN,1.8
    FARMERS-Barcin_Neolithic,1.2
    LEVANT-Neolithic,1
    AFRICA,0.6
    HUNTERS-WHG,0.4
    SOUTH-AMERICA,0.4
    NORTH-AMERICA,0.2




    Last edited by scobar; 07-23-2019 at 09:48 AM.

  8. The Following 5 Users Say Thank You to scobar For This Useful Post:

     26284729292 (07-24-2019),  agent_lime (07-23-2019),  bmoney (07-24-2019),  MonkeyDLuffy (07-23-2019),  pegasus (07-23-2019)

  9. #146
    Registered Users
    Posts
    793
    Sex
    Location
    Netherlands
    Ethnicity
    South-Dutch
    Nationality
    Dutch
    Y-DNA
    I2a2a1b2-CTS1977
    mtDNA
    H13a1a1

    Netherlands Belgium
    Quote Originally Posted by scobar View Post
    Can pen=0 be more useful for mixed individuals or when modeling w/ ancients because it doesn't penalize distant admixtures? You mentioned Davidski and I believe he uses pen=0 on the scaled ancients model he hands out with your G25 co-ords.

    Had another question about the Nbatch parameter too that would be really helpful..I noticed w/ Davidski's scaled ancients model (actually a variation of it using Dmx's aasi sims) that using an nbatch=200 and with the default penalty seemed to consistently produce a non-trivial better fit instead of the default 500 nbatch in nmonte3. The model is just 10 pops or so..I don't know if this is an artifact of the particular model, samples being modeled, or just the size of the model and params being used or something else? Higher vs lower nbatch didn't affect the pen=0 runs much fyi.
    These calculations are heavily stochastic. Sometimes you may find some examples where using a nonstandard parameter returns an interesting result.
    Don't let this fool you into thinking that you have found an improvement of the method.
    In most cases you will be working with overfitted models (especially when you use k=25 unscaled data). In those cases it just stupid to trust a pen=0 model.
    I know that changing the Nbatch parameter may have profound effects on the results. I have tried to understand these effects but I failed. When you publish results of nonstandard Nbatch values, the main effect that is that your results are not comparable with the results of others. So I recommend not to this.

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

     agent_lime (07-24-2019),  MonkeyDLuffy (07-24-2019),  poi (07-24-2019),  scobar (07-24-2019)

  11. #147
    Registered Users
    Posts
    943
    Sex
    Location
    US
    Ethnicity
    India Andhra Kamma Telugu
    Nationality
    US
    Y-DNA
    HM82 Z5888+CTS8144+
    mtDNA
    U2a1a

    United States of America India
    Quote Originally Posted by Censored View Post
    What would a typical Kushan genome(predictably) have looked like? As far as Siberian HG, Sintashta steppe, BMAC, etc?
    https://scienceblogs.com/gnxp/2010/0...d-be-a-barbari

    the peoples that the Chinese termed the Yuezhi and Wusun. A medieval Chinese scholar noted: "Among the various Rong [alien races] in the Western Regions, the Wusun's shape was the strangest; and the present barbarians who have green eyes and red hair, and are like macaques, belonged to the same race as the Wusun." These were likely Indo-European speaking groups, some of whom were affiliated with the Tocharians. How did they get where they were? You can read Empires of the Silk Road or The Horse, the Wheel, and Language for some ideas. Their genetic impact remains evident in China today. The Uyghurs are almost certainly partly descended from them, although they are linguistically a Turkic group.
    The DNA analyses revealed that one subject was an ancient male skeleton with maternal U2e1 and paternal R1a1 haplogroups.


    Chikisheva (2000a) did observe for both male and
    female Pazyryk crania a close biological similarity to the
    ancient Saka and Wusun groups of eastern Kazakhstan and Xinjiang.

    these results point to a high probability of
    sex-biased admixture in both groups where migrating Tagar males assimilated some previous Iron Age population while Pazyryk males are more similar to neighboring Iron Age
    Tuvan peoples who may have had other sources of marital partners
    Y: H-M69 -> H-M82 -> SK1225 -> H-Z5888 -> H-Z5890 -> H-CTS8144 [CTS8144/PF1741/M5498] -> Z34531 (H1a1a4b3b1a8~)
    mtDNA: U2a1a
    extras 309.1C 315.1C 522.1A 522.2 CG8572A G8860A T11368C T16093a T16154C C16519T C195T

    G25 Ancients Dist 0.99 Shahr_I_Sokhta_BA3:S8728.E1.L1 65.2 Saidu_Sharif_IA_o:S7722.E1.L1 17.8 Udegram_IA:I1985 7.8 Jordanian:S_Jordanian-1 4.4 Barikot_IA:I6545 2.2 Scotland_N:I26602 Narva_Lithuania: Donkalnis6

    Lactose persistance rs3213871 rs4988243

  12. #148
    Registered Users
    Posts
    289
    Sex
    Ethnicity
    Malayali Muslim
    Nationality
    Indian
    Y-DNA
    HM82->Z4507->Z5890
    mtDNA
    R30a

    Quote Originally Posted by Huijbregts View Post
    These calculations are heavily stochastic. Sometimes you may find some examples where using a nonstandard parameter returns an interesting result.
    Don't let this fool you into thinking that you have found an improvement of the method.
    In most cases you will be working with overfitted models (especially when you use k=25 unscaled data). In those cases it just stupid to trust a pen=0 model.
    I know that changing the Nbatch parameter may have profound effects on the results. I have tried to understand these effects but I failed. When you publish results of nonstandard Nbatch values, the main effect that is that your results are not comparable with the results of others. So I recommend not to this.
    I did think the Nbatch=250 model was maybe overfitting..just was surprised by the discrepancy being that much between it and the default Nbatch for pen=0.001 runs, so thought maybe there was something here about the Nbatch param and the size/# of samples of the model one's using that I didn't understand.

  13. #149
    Registered Users
    Posts
    943
    Sex
    Location
    US
    Ethnicity
    India Andhra Kamma Telugu
    Nationality
    US
    Y-DNA
    HM82 Z5888+CTS8144+
    mtDNA
    U2a1a

    United States of America India
    Quote Originally Posted by scobar View Post
    I did think the Nbatch=250 model was maybe overfitting..just was surprised by the discrepancy being that much between it and the default Nbatch for pen=0.001 runs, so thought maybe there was something here about the Nbatch param and the size/# of samples of the model one's using that I didn't understand.
    For smaller samples comparing Indians to less than 200 items try around pen=0.009,Ncycles=450,Nbatch=300. check the fit with the GED segment match. Some times it makes sense

  14. The Following User Says Thank You to tipirneni For This Useful Post:

     scobar (07-24-2019)

  15. #150
    Registered Users
    Posts
    289
    Sex
    Ethnicity
    Malayali Muslim
    Nationality
    Indian
    Y-DNA
    HM82->Z4507->Z5890
    mtDNA
    R30a

    Quote Originally Posted by tipirneni View Post
    For smaller samples comparing Indians to less than 200 items try around pen=0.009,Ncycles=450,Nbatch=300. check the fit with the GED segment match. Some times it makes sense
    The sheet was using averages though..

Page 15 of 18 FirstFirst ... 51314151617 ... LastLast

Similar Threads

  1. Replies: 8
    Last Post: 08-01-2019, 05:13 AM
  2. I got my g25 coordinates
    By Magnetic in forum Autosomal (auDNA)
    Replies: 86
    Last Post: 07-02-2019, 07:54 AM
  3. How to determine my AASI amount
    By Censored in forum Southern
    Replies: 8
    Last Post: 08-07-2018, 03:54 AM
  4. Replies: 5
    Last Post: 07-25-2018, 09:39 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
  •