Page 48 of 52 FirstFirst ... 384647484950 ... LastLast
Results 471 to 480 of 520

Thread: The Neolithic Transition in the Baltic Was Not Driven by Admixture with Early Europea

  1. #471
    Registered Users
    Posts
    3,439
    Sex

    Quote Originally Posted by Helgenes50 View Post
    Generalissimo,
    What is the best solution for the NW Europeans, 7 or 10 dimensions ?
    That would require some testing, and I have not done it yet.

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

     Helgenes50 (02-12-2017)

  3. #472
    Banned
    Posts
    4,169
    Sex
    Ethnicity
    N/A
    Nationality
    N/A
    mtDNA (M)
    H 47
    Y-DNA (P)
    I2a1-L621- PH 908

    Quote Originally Posted by Generalissimo View Post
    Models should be based on cemented findings from scientific literature and/or easily reproducible results from formal statistics, using packages like Admixtools.

    Global 10/nMonte or Admixture test of your choice/nMonte can then be used to reproduce and perhaps explore further with extreme caution such models, experimenting with 7-10 dimensions, different samples, and so on. They can't be used to overturn the above or propose some sort of surprising new theories.

    It might sound boring and difficult, but that's the way it is. Doing it any other way is a waste of energy, and it will lead nowhere.
    For my part I use it to probe below what the papers or your models say, and only hard data should be used to make any bold claims. Yet, often even with new data, the academic papers themselves can lend themselves to further analysis.
    After the extensive discussions about dimensions and weighting, I used weighted data, always beginning with a well established dictum as a sanity check.

    For example results like below give some confidence in my set -up:

    Yamnaya_Samara:I0429
    Samara_Eneolithic:I0434 53.35 %
    Latvia_HG:ZVEJ32 21.85 %
    Kotias:KK1 20.7 %
    Ukraine_HG1:StPet2 4.1 %
    Bichon:Bichon 0 %

    Afanasievo:RISE508
    Kotias:KK1 41.15 %
    Samara_HG:I0124 37.2 %
    Latvia_HG:ZVEJ32 21.4 %

    As do the results for CWC & Latvia LN

    Corded_Ware_Germany:I1534
    Yamnaya_Samara:I0231 67.85 %
    Hungary_N:I1495 19.45 %
    Latvia_HG:ZVEJ32 12.45 %
    Kotias:KK1 0.25 %

    Latvia LN
    Latvia_HG:ZVEJ32 44.25 %
    Kotias:KK1 35.6 %
    Samara_Eneolithic:I0434 20 %

    A few percentage points differing to yours is to expected and not alarming .

    So one does wonder about some results in which the "steppe ancestry" for some post-3000 BC Europeans doesn't appear to derive directly from Yamnaya itself, the st up is otherwise the same, and covergae isn't an issue. Are they to be discarded or are they true outliers ?
    We should also be aware in that Admixtools & such isn't necessarily a perfect system either, when it models the Neolithic ancestry of Spaniards or Basques as exclusively coming from Lengyel (what is obviously incorrect or a limitation of the run).
    Last edited by Gravetto-Danubian; 02-12-2017 at 09:36 AM.

  4. The Following User Says Thank You to Gravetto-Danubian For This Useful Post:

     Silesian (02-13-2017)

  5. #473
    Registered Users
    Posts
    530
    Sex
    Ethnicity
    Dutch
    mtDNA (M)
    HV
    Y-DNA (P)
    R1b-Z326>A5869

    Quote Originally Posted by Gravetto-Danubian View Post
    Davidski got similar results with qpAdm, so its maybe something Radboud is doing.
    What do you get for NW European populations like German and South_East English?

    Quote Originally Posted by Kale View Post
    Was this addressing me? If so yes, that's absolutely a wild claim, I was pointing out how Radboud's analysis makes no sense.
    It was an experiment, no need to point out the results dont make sense.

  6. #474
    Banned
    Posts
    4,169
    Sex
    Ethnicity
    N/A
    Nationality
    N/A
    mtDNA (M)
    H 47
    Y-DNA (P)
    I2a1-L621- PH 908

    Quote Originally Posted by Radboud View Post
    What do you get for NW European populations like German and South_East English?

    German
    Latvia_HG:ZVEJ32 31.2 %
    Hungary_N:I1495 25.9 %
    Yamnaya_Samara:I0231 19.65 %
    Armenia_EBA:I1635 15.55 %
    Baalberge_MN:I0560 5.8 %

    English_Kent
    Latvia_HG:ZVEJ32 34.15 %
    Hungary_N:I1495 30.25 %
    Armenia_EBA:I1635 23.3 %
    Yamnaya_Samara:I0231 12.2 %

    Norwegian
    Yamnaya_Samara:I0231 37.45 %
    Hungary_N:I1495 20.7 %
    Latvia_HG:ZVEJ32 20.7 %
    Baalberg_MN: 21.15 %

    Irish
    Latvia_HG:ZVEJ32 28.5 %
    Yamnaya_Samara:I0231 22.9 %
    Hungary_N:I1495 20.05 %
    Armenia_EBA:I1635 17.35 %
    Baalberge_MN:I0560 11.2 %

    The main source of extra WHG is like LatviaHG (ie the Baltic), reccurringly; the main steppe source is Yamnaya, EEF is from Lengyel & MN Germany, and some Armenia EBA (? historic impacts) comes into play.
    You can then try to differentiate between CWC, BB , post Beaker sources of steppe, but that gets more shakey. And obviously models will shift as we acquire more relevant samples for a region.

    It was an experiment, no need to point out the results dont make sense.
    I agree, we can make more gentle critiques. We all begin somewhere, no one is omniscient, and have even at times posted chronologically non-viable results
    Last edited by Gravetto-Danubian; 02-12-2017 at 08:59 AM.

  7. The Following 5 Users Say Thank You to Gravetto-Danubian For This Useful Post:

     pegasus (02-12-2017),  peternorth (02-12-2017),  Radboud (02-12-2017),  rms2 (02-12-2017),  Silesian (02-13-2017)

  8. #475
    Banned
    Posts
    5,173
    Sex

    Quote Originally Posted by wombatofthenorth View Post
    HMm I don't know. Other things suggested otherwise. Maybe they got their input later on? Or in some different way. Whatever the case they did not seem to come under the control of the Steppe people who brought Indo-European languages to all of the rest of Europe.

    They just happen to be on a diminishing NE-SW cline for steppe ancestry in Europe generally, and Spain specifically...probably a little more interaction with their European neighbors to the NE than the rest of Spain.Also, SW Spain probably has more African related admixture which would tend to decrease shared drift stats with steppe a little.

  9. The Following User Says Thank You to Kurd For This Useful Post:

     ArmandoR1b (02-12-2017)

  10. #476
    Registered Users
    Posts
    1,359
    Sex
    Omitted

    Quote Originally Posted by Gravetto-Danubian View Post
    I agree, we can make more gentle critiques. We all begin somewhere, no one is omniscient, and have even at times posted chronologically non-viable results
    Can't get it over it can ya? We've been over this already

    EDIT: Here's what I get for English_Kent
    88.2% Bell_Beaker_Germany
    3.4% Iberia_MN
    3.2% Iberia_Chalcolithic
    1.9% Loschbour
    1.8% Armenia_EBA
    And some noise
    0.9% Ulchi
    0.6% Yoruba
    Outgroups: Ami, Anatolia_Neolithic, Bichon, Iberia_EN, Iran_Chalcolithic, Karitiana, Kostenki14, Kotias, Papuan, Ust_Ishim, Yamnaya_Samara
    Last edited by Kale; 02-12-2017 at 06:05 PM.

  11. The Following User Says Thank You to Kale For This Useful Post:

     Gravetto-Danubian (02-12-2017)

  12. #477
    Registered Users
    Posts
    1,250
    Sex
    Location
    USA
    Ethnicity
    Latvian
    Nationality
    USA
    mtDNA (M)
    k1a4a1
    Y-DNA (P)
    R-L20

    Quote Originally Posted by lgmayka View Post
    Perhaps your ancestor fought in the Battle of Grunwald, on the side of the Poles-Lithuanians-Belarusians-Ukrainians-Czechs-Romanians-Tatars-etc, then decided to stay in the Baltic region instead of returning to Bohemia or Silesia?
    Yeah, I don't know, could be. It will be interesting to see if the STR67 upgrade shows a closer or more distant match to the distance 3 guy at STR37.
    At some point our specific branch of the line ended up become treated as surnameless Latvian serfs for at least a couple generations and maybe many more for all I know.

  13. #478
    Quote Originally Posted by ArmandoR1b View Post
    I wish people would stop using that useless call for PF6518, which is only a phylogenetic equivalent of M269 and not a synonym, but useless no matter what.




    http://www.anthrogenica.com/showthre...l=1#post160114


    I knew that PF6518, a phylogenetic equivalent of M269, was an unreliable call from the very beginning.
    Quote Originally Posted by R.Rocca View Post
    And while we are at it, the Atapuerca ATP3 PF6518+ (M269 equivalent) call is pretty useless as well. Here are the same random samples I used for I0443:

    Atapuerca ATP3 - Haplogroup R - 1 of 1 derived reads
    L2_Rocca - Haplogroup R - 20 of 20 derived reads
    NA19309 - Haplogroup E - 2 of 2 derived reads
    NA18868 - Haplogroup E - 0 of 4 derived reads
    NA12286 - Haplogroup G - 4 of 6 derived reads
    NA06994 - Haplogroup I - 2 of 5 derived reads
    NA18611 - Haplogroup O - 1 of 1 derived reads



    The positive call for R1b1a-FGC46/Y97 is legitimate though.
    Quote Originally Posted by R.Rocca View Post
    On ATP3?... It was never solid because there are no reads for M269. So the last reliable read is R1b1a-FGC46/Y97.

    to get an idea of the number of reads and quality of the data, here is a compare in BAM file sizes...

    139,610,336 KB... L2_Rocca (Whole Genome)
    682,353 KB... I0443 (Whole Genome)
    83,060 KB... ATP3 (Whole Genome)

    So it is pretty telling why the authors chose not to call the haplogroup for ATP3 or why nobody uses it to plot it on a PCR.
    These claims made by Richard Rocca about PF6518 being "pretty useless", which ArmandoR1b has so enthusiastically repeated, and which have led to other commenters on other websites claiming that ATP3 was only R1b1a1a-P297 and not R1b1a1a2-M269, are completely false.

    The reason that Rocca observed the derived allele of PF6518 for 1000 Genomes samples not belonging to R1b is that he was looking at the phase 1 data for those samples, which consisted of short reads with low mapping quality (less than 30). What happened is that similar sequences from other parts of the genome were incorrectly aligned to the position of PF6518 on the Y chromosome, making it seem as though the non-R1b samples had the derived allele of PF6518, when in fact they didn't.

    The phase 3 data for the 1000 Genomes samples consists of long reads with high mapping quality (greater than 30), and if you look at that data for the same non-R1b samples that Rocca looked at, you'll see that they all have only the ancestral allele of PF6518.

    The read for ATP3 that's aligned to PF6518 is a long read (109 bases) with a high mapping quality (37), so the positive call for PF6518 for ATP3 is completely reliable.

    Quote Originally Posted by ArmandoR1b View Post
    SAMEA3477467 ATP3 only has 81 MB of coverage which is extremely poor. The ONLY M269 SNP that Genetiker was able to find was PF6518 which is found in many other haplogroups. Please do a search of it at http://www.yfull.com/search-snp-in-tree/ That same specimen also has calls for SNPs belonging to other haplogroups yet Genetiker only pointed out PF6518 which lead him to M269. That's being very careless.

    Here is the result of the search of PF6518 at YFull.

    − PF6518 G-FGC5672
    This position for SNP is not in the YTree
    − PF6518 J-CTS130
    This position for SNP is not in the YTree
    − PF6518 E-CTS1622
    This position for SNP is not in the YTree
    − PF6518 C-V20*
    This position for SNP is not in the YTree
    − PF6518 C1a1a1
    This position for SNP is not in the YTree
    − PF6518 R-M269
    − PF6518 N-Y4374
    This position for SNP is not in the YTree
    Do the search again. You'll see that the derived allele occurs only in C-V20*, G-FGC5672*, and R1b-M269. C-V20* is La Brańa 1 and G-FGC5672* is Ötzi. The apparent presence of the derived allele in those two ancient samples is due either to misalignment or DNA damage.

  14. The Following 13 Users Say Thank You to genetiker For This Useful Post:

     bicicleur (02-13-2017),  Genarh (02-13-2017),  Isidro (02-13-2017),  lgmayka (02-14-2017),  Megalophias (02-13-2017),  palamede (02-13-2017),  parasar (02-13-2017),  Pribislav (02-13-2017),  Romilius (02-13-2017),  Saetro (02-13-2017),  Silesian (02-13-2017),  venustas (02-13-2017),  vettor (02-13-2017)

  15. #479
    Registered Users
    Posts
    767
    Sex
    Ethnicity
    Italian

    Italy
    Quote Originally Posted by wombatofthenorth View Post
    Yeah, I don't know, could be. It will be interesting to see if the STR67 upgrade shows a closer or more distant match to the distance 3 guy at STR37.
    At some point our specific branch of the line ended up become treated as surnameless Latvian serfs for at least a couple generations and maybe many more for all I know.
    I don't think that evoking serfdom is useful... probably, you didn't search properly: surnames are likely to change many times in past, before the creation of a real and institutional registrar of French derivation (for example, in Italy, surnames ceased to change and, in those region where they didn't exist at all, like in some Umbria valleys, to be fixed only AFTER the arrival of Napoleon, who creates a registrar. The previous registrar weren't precise as French one, in the sense that, as local, they followed the manner people called other people, so a nickname could easily be registered in place of the real surname and, in the next year census, the nickname remained as surname). Probably, in Latvia happened more or less the same. I think you should try to see the records of German places in your region: Hanseatic League was present... I remember well when I went to Lithuania, Latvia and Estonia a lot of German Hansa houses and also place names! Cross results of people with the same family name and follow collateral branches when your line is difficult.

  16. The Following User Says Thank You to Romilius For This Useful Post:

     palamede (02-13-2017)

  17. #480
    Registered Users
    Posts
    767
    Sex
    Ethnicity
    Italian

    Italy
    Quote Originally Posted by genetiker View Post
    These claims made by Richard Rocca about PF6518 being "pretty useless", which ArmandoR1b has so enthusiastically repeated, and which have led to other commenters on other websites claiming that ATP3 was only R1b1a1a-P297 and not R1b1a1a2-M269, are completely false.

    The reason that Rocca observed the derived allele of PF6518 for 1000 Genomes samples not belonging to R1b is that he was looking at the phase 1 data for those samples, which consisted of short reads with low mapping quality (less than 30). What happened is that similar sequences from other parts of the genome were incorrectly aligned to the position of PF6518 on the Y chromosome, making it seem as though the non-R1b samples had the derived allele of PF6518, when in fact they didn't.

    The phase 3 data for the 1000 Genomes samples consists of long reads with high mapping quality (greater than 30), and if you look at that data for the same non-R1b samples that Rocca looked at, you'll see that they all have only the ancestral allele of PF6518.

    The read for ATP3 that's aligned to PF6518 is a long read (109 bases) with a high mapping quality (37), so the positive call for PF6518 for ATP3 is completely reliable.



    Do the search again. You'll see that the derived allele occurs only in C-V20*, G-FGC5672*, and R1b-M269. C-V20* is La Brańa 1 and G-FGC5672* is Ötzi. The apparent presence of the derived allele in those two ancient samples is due either to misalignment or DNA damage.
    I see, thanks for sharing... I think that ATP3 is really strange... You also stated he's got a lot of eastern admixture... do you think he is an early arrival in Iberia from the East?

Page 48 of 52 FirstFirst ... 384647484950 ... LastLast

Similar Threads

  1. Neolithic Near East ADMIXTURE Results
    By Kurd in forum Autosomal (auDNA)
    Replies: 96
    Last Post: 06-27-2016, 05:39 AM
  2. Replies: 9
    Last Post: 11-30-2015, 08:48 PM
  3. Did Neolithic Iberians have any Negroid admixture?
    By One-Eyed King in forum Ancient (aDNA)
    Replies: 4
    Last Post: 10-16-2015, 03:51 AM
  4. Early Fertile Crescent Neolithic Y-DNA
    By J Man in forum Ancient (aDNA)
    Replies: 3
    Last Post: 02-26-2015, 11:38 PM
  5. Replies: 6
    Last Post: 04-18-2013, 08:52 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
  •