Page 696 of 744 FirstFirst ... 196596646686694695696697698706 ... LastLast
Results 6,951 to 6,960 of 7440

Thread: Could Western Jews (Ash. and Seph.) descend from Aegeans and Levantine admixture?

  1. #6951
    Registered Users
    Posts
    852
    Sex
    Ethnicity
    Ashkenazi Jew
    Nationality
    American
    Y-DNA (P)
    Q-Y2750
    mtDNA (M)
    H47

    United States of America Israel
    Quote Originally Posted by Bleach View Post
    IMO, It's a typical case of overfitting here
    In fact, the single distance between Levant LBN Roman and modern Lebanese christian is only 1,97, well you use 8 additional references
    to get to 1,13. In fact we could have even called it quits at 1,97
    I removed Levant LBN Roman and kept the other 8 samples



    As you can see, it creates a synthetic modern Lebanese Christian with a very good fit but totally fake from a historical timescale
    Actually, Bedouins+ Egyptians harbor the ancient Natufian background , Aegan BA + Anatolian BA harbor Anatolian Farmers input over the Neolithic and Iran IA habor the Iranian
    farmers input over the chalcocithic
    So in blending that synthetic Levantine with Roman era Levantines you'll certainly obtain very good fits but it will be impossible
    to detect the actual Roman era Levantine contribution which is likely close to 90% for Lebanese Christians IMO.
    Yes, but the exact same argument can be made for your models of Western Jews. You're assuming large numbers of converts when there's no direct evidence for this. We know that there were converts, we don't know the proportion. So by creating a model with all of these populations for Western Jews, you can easily reduce the Levantine component just as you can for modern Levantines who we know are primarily descended from the locals.

    You can model Western Jews with Cypriots and reduce the Levantine to almost nothing, but we know that's ridiculous. Yet I've actually seen people online claim that many Western Jews are simply Cypriots with additional North African admixture.

    And I agree that Lebanese Christians are mostly descended from Roman era Levantines... Just as I believe that Western Jews are 40-50% descended from Roman era Levantines.
    Last edited by jonahst; 08-13-2019 at 05:53 PM.

  2. The Following 7 Users Say Thank You to jonahst For This Useful Post:

     Agamemnon (08-13-2019),  Andrewid (08-13-2019),  Bleach (08-13-2019),  Erikl86 (08-13-2019),  hartaisarlag (08-13-2019),  Power77 (08-13-2019),  StillWater (08-13-2019)

  3. #6952
    Registered Users
    Posts
    852
    Sex
    Ethnicity
    Ashkenazi Jew
    Nationality
    American
    Y-DNA (P)
    Q-Y2750
    mtDNA (M)
    H47

    United States of America Israel
    Quote Originally Posted by StillWater View Post
    I'm not singling anyone out, but it's about time this is pointed out, assuming it hasn't been already. When these models get beyond a couple variables, it seems that they only become good at ordering populations by reference category instead of actually measuring that category. This is clearly seen when two very close Jewish groups are compared or here in Jonah's model, where Lebanese Muslims have about 4x a category of Lebanese Christians. This is fine when it's 4x noise, but that isn't the case with these models. I'm not singling Jonah out; I'm just too lazy to go back to previous pages for more examples. This happens almost every time. The same occurs when I make these models. I can't think of anyone who is an exception. In the case of Jonah's model, for example, we can say that Lebanese Muslims have more ARM LBA Averaged than Lebanese Christians, but it's silly to take the absolute numbers or even their neighborhood seriously at this point. Maybe when we understand G25 and the nMonte function better, then we can take the numbers as rough measurements. But for now, I recommend we simply use them to order populations.
    I believe I agree with you. My entire point in creating the models for modern Levantines was to show that Bleach's models for Western Jews are misleading.
    Last edited by jonahst; 08-13-2019 at 03:56 PM.

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

     Agamemnon (08-13-2019),  Erikl86 (08-13-2019),  Power77 (08-13-2019),  StillWater (08-13-2019)

  5. #6953
    Registered Users
    Posts
    137
    Sex
    Location
    Paris
    Y-DNA (P)
    J-CTS5368
    mtDNA (M)
    H3c1

    Quote Originally Posted by jonahst View Post
    Yes, but the exact same argument can be made for your models of Western Jews. You're assuming large numbers of converts when there's no direct evidence for this. We know that there were converts, we don't know the proportion. So by creating a model with all of these populations for Western Jews, you can easily reduce the Levantine component just as you can for modern Levantines who we know are primarily descended from the locals.

    OK, but we know the native populations who brought converts : Hellenistic Egyptians , Aegeans, Anatolians, etc…
    That being known, it’s the nmonte’ s job to determine the ancestry proportion as we put the related references in the set
    If not, what the point of nmonte ?
    As to your Lebanese models I would add that we also know that Bedouins and Egyptian did not contribute to their ethnogenesis , so in that case it’s clearly a mistake to include these references in the set but It's not a mistake to include Egyptians, Anatolians, etc in the western Jews nmonte set since we know they were an historical input

    Quote Originally Posted by jonahst View Post
    You can model Western Jews with Cypriots and reduce the Levantine to almost nothing, but we know that's ridiculous. Yet I've actually seen people online claim that many Western Jews are simply Cypriots with additional North African admixture.

    And I agree that Lebanese Christians are mostly descended from Roman era Levantines... Just as I believe that Western Jews are 40-50% descended from Roman era Levantines.
    OK Let's suppose that fictional situation :
    So I'm 1/4 Lebanese christian, 1/4 Assyrian (Christian),1/4 Cypriot (without any Maronite ancestry),1/4 Sicilian
    and that all these four backgrounds are proven.
    I decide to run my sample on the G25 nmonte
    First, I pick these 4 reference samples and I obtain fairly 25% of each of them with a distance of 1
    Then , I pick Bergamo, Sicilian,Kurdish, Lebanese , and I obtain 50% Lebanese,25% Sicilian,12,5% Bergamo and 12,5% Kurdish
    with as well a distance of 1

    Is it because I love my Lebanese background , and anyway the distance is the same than with the 4 real ancestries that I’m allowed to not include Cypriots and Assyrians in the set?

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

     Andrewid (08-13-2019),  Claudio (08-13-2019)

  7. #6954
    Registered Users
    Posts
    852
    Sex
    Ethnicity
    Ashkenazi Jew
    Nationality
    American
    Y-DNA (P)
    Q-Y2750
    mtDNA (M)
    H47

    United States of America Israel
    Quote Originally Posted by Bleach View Post
    OK, but we know the native populations who brought converts : Hellenistic Egyptians , Aegeans, Anatolians, etc…
    That being known, it’s the nmonte’ s job to determine the ancestry proportion as we put the related references in the set
    If not, what the point of nmonte ?
    As to your Lebanese models I would add that we also know that Bedouins and Egyptian did not contribute to their ethnogenesis , so in that case it’s clearly a mistake to include these references in the set but It's not a mistake to include Egyptians, Anatolians, etc in the western Jews nmonte set since we know they were an historical input


    OK Let's suppose that fictional situation :
    So I'm 1/4 Lebanese christian, 1/4 Assyrian (Christian),1/4 Cypriot (without any Maronite ancestry),1/4 Sicilian
    and that all these four backgrounds are proven.
    I decide to run my sample on the G25 nmonte
    First, I pick these 4 reference samples and I obtain fairly 25% of each of them with a distance of 1
    Then , I pick Bergamo, Sicilian,Kurdish, Lebanese , and I obtain 50% Lebanese,25% Sicilian,12,5% Bergamo and 12,5% Kurdish
    with as well a distance of 1

    Is it because I love my Lebanese background , and anyway the distance is the same than with the 4 real ancestries that I’m allowed to not include Cypriots and Assyrians in the set?
    Yes, we know there were converts, we don't know how many. The degree of such admixture will vary considerably based on what populations you include. You're assuming that each nearby regions, regions that can also easily be used to model modern Levantines, contributed a significant enough portion of ancestry to warrant inclusion.

    Historical accounts exaggerate and can be misleading. If we were going only be occasional and often vague historical accounts, we could attribute significance to the Khazar narrative. Or, based on population growth (such as citing some ancient claims about what % of the Roman Empire was Jewish) that significant Eastern European ancestry was necessary for the "demographic miracle" of Ashkenazi Jews in Eastern Europe.

    In terms of single population distances, Western Jews fall along of a SE to NW cline, having East Med as their closest populations followed by Levantines and Southern Europeans.

    Modeling Samaritans similarly (but with BA Levantines) will generally not include such high Egyptian and Bedouin. Do we know definitively that Copts and Ghassanids didn't migrate into Lebanon along with Anatolians and Mesopotamians. To be clear, I do not think this narrative has any meaningful basis, but my point is that it's not difficult to support an exaggerated historical narrative using genetic evidence if you want to.

  8. The Following 6 Users Say Thank You to jonahst For This Useful Post:

     Andrewid (08-13-2019),  Bleach (08-13-2019),  Claudio (08-13-2019),  Erikl86 (08-13-2019),  Power77 (08-13-2019),  StillWater (08-13-2019)

  9. #6955
    Registered Users
    Posts
    265
    Sex
    Ethnicity
    95% Ashkenaz,5% Sephardic
    Nationality
    US
    Y-DNA (P)
    J2a
    mtDNA (M)
    J1c

    Have not read the Yardanian and Schurr full texts. However they are mainly experts on the Svans of the Caucasus. In a minor way, so am I. Jews, for the most part, did not descend from the Svans or any other Caucasian groups . Jewish Y dna J2a1b1 lines are cousins of the Caucasian and Svan lines, but Jewsh lines are not downstream of those lines. According to the J2a Project, J2a1b1 lines can be separated into Caucasian and non Caucasian lines. The Jewish subclades have mutation Z5048 and the Caucasian subclades do not. Most likely, the lines separated near the Zagros mountains since there are Punjabi Z5048 lines. At the Project, my mainly Jewish subclade is separated from the Caaucasian subclades despite the overlap. (I have corresponded with a member of the Svan group) I also would hope for more scholarly work

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

     Andrewid (08-13-2019),  Erikl86 (08-13-2019),  jonahst (08-14-2019),  Power77 (08-13-2019)

  11. #6956
    Registered Users
    Posts
    1,164

    Quote Originally Posted by jonahst View Post
    Yes, but the exact same argument can be made for your models of Western Jews. You're assuming large numbers of converts when there's no direct evidence for this. We know that there were converts, we don't know the proportion. So by creating a model with all of these populations for Western Jews, you can easily reduce the Levantine component just as you can for modern Levantines who we know are primarily descended from the locals.

    You can model Western Jews with Cypriots and reduce the Levantine to almost nothing, but we know that's ridiculous. Yet I've actually seen people online claim that many Western Jews are simply Cypriots with additional North African admixture.

    And I agree that Lebanese Christians are mostly descended from Roman era Levantines... Just as I believe that Western Jews are 40-50% descended from Roman era Levantines.
    For me Jonah it all depends on what ethogenesis narrative and timescale we are going with for western Jews?

    For Ashkenazim we have one narrative of post Heraclius revolt 7th Century full Levantine Jews moving to North Italy temporarily for 200 years then moving France/Germany then Eastern Europe.
    So whole Narrative lasts from full Levantine population in Levant 700AD to 1500 Eastern Europe so Roughly a genetic timescale narrative of only 700 years.

    2nd Narrative is 1st Century AD already partially mixed Levantine population movements plus a stay in Southern Europe for near on 600 years plus another 500 years bringing us up to Eastern Europe in 1500’s So in comparison we are talking about a partially mixed Levantine population with a genetic timescale narrative of twice as long as the first narrative at Roughly 1400 years.

    With both narratives we know there were converts to some degree (evidently almost exclusively women (Southern European,then Western and Eastern European women.

    With the first narrative I can see the possibility as you mention that of Western Jews being 40-50% Autosomaly descended from Roman era Levantines.

    I’m just not 100% convinced it would be possible with the second narrative which is for starters twice as long as the first,snd starts with an already admixed Levantine population movement.
    Plus then we have to factor in that we don’t know how much modern Autosomal Admixture of Western Jews was picked up solely as maternally inherited Admixture during this Roman period and subsequent extra 600 years in Southern Europe so Western Jews coming up only 25/30% Autosomaly Roman era Levantines doesn’t necessarily seem that unreasonable all things considered.
    Last edited by Claudio; 08-13-2019 at 08:44 PM.

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

     Andrewid (08-13-2019),  Bleach (08-13-2019),  jonahst (08-14-2019)

  13. #6957
    Registered Users
    Posts
    265
    Sex
    Ethnicity
    95% Ashkenaz,5% Sephardic
    Nationality
    US
    Y-DNA (P)
    J2a
    mtDNA (M)
    J1c

    Quote Originally Posted by josh w. View Post
    Have not read the Yardanian and Schurr full texts. However they are mainly experts on the Svans of the Caucasus. In a minor way, so am I. Jews, for the most part, did not descend from the Svans or any other Caucasian groups . Jewish Y dna J2a1b1 lines are cousins of the Caucasian and Svan lines, but Jewsh lines are not downstream of those lines. According to the J2a Project, J2a1b1 lines can be separated into Caucasian and non Caucasian lines. The Jewish subclades have mutation Z5048 and the Caucasian subclades do not. Most likely, the lines separated near the Zagros mountains since there are Punjabi Z5048 lines. At the Project, my mainly Jewish subclade is separated from the Caaucasian subclades despite the overlap. (I have corresponded with a member of the Svan group) I also would hope for more scholarly work
    To clarify, my subclade is Western Jewish and not Mizrachi

  14. The Following 2 Users Say Thank You to josh w. For This Useful Post:

     Erikl86 (08-13-2019),  Power77 (08-13-2019)

  15. #6958
    Registered Users
    Posts
    137
    Sex
    Location
    Paris
    Y-DNA (P)
    J-CTS5368
    mtDNA (M)
    H3c1

    Quote Originally Posted by jonahst View Post
    Yes, we know there were converts, we don't know how many. The degree of such admixture will vary considerably based on what populations you include. You're assuming that each nearby regions, regions that can also easily be used to model modern Levantines, contributed a significant enough portion of ancestry to warrant inclusion.

    Historical accounts exaggerate and can be misleading. If we were going only be occasional and often vague historical accounts, we could attribute significance to the Khazar narrative. Or, based on population growth (such as citing some ancient claims about what % of the Roman Empire was Jewish) that significant Eastern European ancestry was necessary for the "demographic miracle" of Ashkenazi Jews in Eastern Europe.

    In terms of single population distances, Western Jews fall along of a SE to NW cline, having East Med as their closest populations followed by Levantines and Southern Europeans.

    Modeling Samaritans similarly (but with BA Levantines) will generally not include such high Egyptian and Bedouin. Do we know definitively that Copts and Ghassanids didn't migrate into Lebanon along with Anatolians and Mesopotamians. To be clear, I do not think this narrative has any meaningful basis, but my point is that it's not difficult to support an exaggerated historical narrative using genetic evidence if you want to.
    Fair enough,
    But that way of reasonning brings about the inverse effect..
    If we can no longer use the closest populations from Levantines in the nmonte set, like Egyptians and Anatolians, because they could potentially
    tap Levantine ancestry through their combination, then you are bound to attribute a quite portion of your 40% -50% Levant LBN Roman to
    each population of converts that could be modelized with Levant LBN as Southern Italians , Anatolians, Egyptians and even Berbers
    Finally, it's tomayto, tomahto isn't it ?

  16. #6959
    Registered Users
    Posts
    668
    Ethnicity
    Jewish

    Israel Jerusalem
    Quote Originally Posted by Bleach View Post
    OK, but we know the native populations who brought converts : Hellenistic Egyptians , Aegeans, Anatolians, etc…
    That being known, it’s the nmonte’ s job to determine the ancestry proportion as we put the related references in the set
    If not, what the point of nmonte ?
    As to your Lebanese models I would add that we also know that Bedouins and Egyptian did not contribute to their ethnogenesis , so in that case it’s clearly a mistake to include these references in the set but It's not a mistake to include Egyptians, Anatolians, etc in the western Jews nmonte set since we know they were an historical input


    OK Let's suppose that fictional situation :
    So I'm 1/4 Lebanese christian, 1/4 Assyrian (Christian),1/4 Cypriot (without any Maronite ancestry),1/4 Sicilian
    and that all these four backgrounds are proven.
    I decide to run my sample on the G25 nmonte
    First, I pick these 4 reference samples and I obtain fairly 25% of each of them with a distance of 1
    Then , I pick Bergamo, Sicilian,Kurdish, Lebanese , and I obtain 50% Lebanese,25% Sicilian,12,5% Bergamo and 12,5% Kurdish
    with as well a distance of 1

    Is it because I love my Lebanese background , and anyway the distance is the same than with the 4 real ancestries that I’m allowed to not include Cypriots and Assyrians in the set?
    I know that Ashkenazim should have background from the following: Lebanese Christians(if you argue that a proxy changes everything, then how many models do you have to toss out?), Tuscans, and Poles. G25 yields this:

    "sample": "Ashkenazi_Jew:Average",
    "fit": 1.5512,
    "Lebanese_Christian": 50.83,
    "Italian_Tuscan": 40,
    "Polish": 9.17,

    I then read a bit more and learn that Ashkenazim should have German ancestry too:

    "sample": "Ashkenazi_Jew:Average",
    "fit": 1.5108,
    "Lebanese_Christian": 55.83,
    "Italian_Tuscan": 22.5,
    "German": 21.67,
    "Polish": 0,

    Looks like there is no Polish ancestry after all.
    Last edited by StillWater; 08-13-2019 at 08:47 PM. Reason: clarity
    הִנְנִי֩ מֵבִ֨יא אוֹתָ֜ם מֵאֶ֣רֶץ צָפ֗וֹן

    Jeremiah 31

  17. The Following 2 Users Say Thank You to StillWater For This Useful Post:

     Bleach (08-13-2019),  Power77 (08-13-2019)

  18. #6960
    Registered Users
    Posts
    137
    Sex
    Location
    Paris
    Y-DNA (P)
    J-CTS5368
    mtDNA (M)
    H3c1

    Quote Originally Posted by StillWater View Post
    I know that Ashkenazim should have background from the following: Lebanese Christians(if you argue that a proxy changes everything, then how many models do you have to toss out?), Tuscans, and Poles. G25 gives me this:

    "sample": "Ashkenazi_Jew:Average",
    "fit": 1.5512,
    "Lebanese_Christian": 50.83,
    "Italian_Tuscan": 40,
    "Polish": 9.17,

    I then read a bit more and learn that Ashkenazim should have German ancestry too:

    "sample": "Ashkenazi_Jew:Average",
    "fit": 1.5108,
    "Lebanese_Christian": 55.83,
    "Italian_Tuscan": 22.5,
    "German": 21.67,
    "Polish": 0,

    Looks like there is no Polish ancestry after all.
    I really dont know how explaining that
    Maybe are you an immortal descendant of a Judean warrior and a Tuscan milkmaid ?
    Your story reminds me the fate of this guy

  19. The Following 3 Users Say Thank You to Bleach For This Useful Post:

     Claudio (08-13-2019),  Power77 (08-13-2019),  StillWater (08-13-2019)

Page 696 of 744 FirstFirst ... 196596646686694695696697698706 ... LastLast

Similar Threads

  1. Replies: 19
    Last Post: 12-07-2018, 12:52 AM
  2. Present-day Lebanese descend from Biblical Canaanites
    By MikeWhalen in forum Ancient (aDNA)
    Replies: 0
    Last Post: 07-27-2017, 07:49 PM
  3. Early farmers from across Europe were direct descendants of Aegeans
    By rock hunter in forum Archaeology (Prehistory)
    Replies: 0
    Last Post: 06-21-2016, 10:43 PM
  4. Replies: 0
    Last Post: 06-14-2016, 01:02 PM
  5. Replies: 0
    Last Post: 05-26-2015, 05:29 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
  •