Page 58 of 74 FirstFirst ... 848565758596068 ... LastLast
Results 571 to 580 of 734

Thread: Genetic origins of the Minoans and Mycenaeans

  1. #571
    Registered Users
    Posts
    3,665
    Sex
    Location
    Australia
    Ethnicity
    Italian Alpine
    Nationality
    Australian and Italian
    Y-DNA
    T1a2b- Z19945
    mtDNA
    H95a- Ostrogoth

    Australia Italy Veneto Friuli Italy Trentino Alto Adige Italy Sweden Finns
    IMO, low-German is the original german , based in north-germany and denmark ..............even the old frisians who lived from denmark to netherlands spoke an old germanic tongue .

    middle-german and high-german dialects came after low-german

    I just realised that people are commenting on linguistics in the thread .......strange
    Last edited by vettor; 08-07-2017 at 07:24 PM.

    European - 99.2%............Central Asian - 0.8% .............Yfull - 1460BC
    Father's Mtdna .....T2b17.......1735 Porcellengo Veneto Italy
    Sons Mtdna .....K1a4 ...........1710 Carnic Alps

    My Path = ( K-M9+, TL-P326+, T-M184+, L490+, M70+, PF5664+, L131+, L446+, CTS933+, CTS54+, CTS8862+, Z19945+, A339+ )

    The main negatives = ( M193-, P322-, P327-, Pages11- , L25- , CTS1848- )

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

     Hando (09-01-2017)

  3. #572
    Gold Class Member
    Posts
    7,300
    Sex
    Location
    United Kingdom
    Nationality
    British
    mtDNA
    H

    United Kingdom
    Folks - the Origins of Germanic thread is this way : http://www.anthrogenica.com/showthre...ns-of-Germanic

    We've been a bit derailed here.

  4. The Following 9 Users Say Thank You to Jean M For This Useful Post:

     Agamemnon (08-07-2017), ArmandoR1b (08-07-2017), Awale (08-08-2017), ffoucart (08-08-2017), Hando (09-01-2017), jdean (08-07-2017), Power77 (08-08-2017), rms2 (08-07-2017), vettor (08-08-2017)

  5. #573
    Moderator
    Posts
    1,398
    Sex
    Location
    Warsaw, Poland
    Y-DNA
    R1a-L1280
    mtDNA
    H2a2(b1)

    Poland European Union
    Quote Originally Posted by Jean M View Post
    Non-linguists messing around just with lexicon in a computer get the thumbs down from me, though naturally I was pleased that Chang et al 2015 failed to replicate the idiotic dates of Bouckaert et al. 2012 and the rest of the Gray and Atkinson school, while sticking to the same methodology.
    I don't think your above remark is correct. In fact, Chang et al. did replicate the dates of Bouckaert et al. See their models from series B in Fig. 5 (especially model B1) and compare it with Fig. 3. Also, please take a look at Fig. 6 where Chang et al. present their results for model C3, in which Archaic PIE (Proto-Indo-Hittite) is dated to about 7700 BP, Early PIE (Proto-Indo-Tocharian) is dated to about 7000 BP, and Late PIE is dated to about 6400 BP. On the other hand, they have also shown that changing the methodology a bit produces significantly younger dates for PIE, and they consider those modified models (like A1 or A2) more reliable.
    Last edited by Michał; 08-08-2017 at 12:14 AM.

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

     Coldmountains (08-08-2017), ffoucart (08-08-2017), Hando (09-01-2017), Power77 (08-08-2017)

  7. #574
    Gold Class Member
    Posts
    7,300
    Sex
    Location
    United Kingdom
    Nationality
    British
    mtDNA
    H

    United Kingdom
    Quote Originally Posted by Michał View Post
    On the other hand, they have also shown that changing the methodology a bit produces significantly younger dates for PIE, and they consider those modified models (like A1 or A2) more reliable.
    Yes I had another look at it after posting that. I suspect that Garrett was responsible for the ancestry constraints which improved the methodology from the completely fatuous to the merely risible.

  8. The Following 2 Users Say Thank You to Jean M For This Useful Post:

     Hando (09-01-2017), vettor (08-08-2017)

  9. #575
    Gold Member Class
    Posts
    547
    Sex
    Location
    Gulf of Fars
    Ethnicity
    Somali
    Y-DNA
    E-V32
    mtDNA
    N1a

    Somalia Ethiopia Eritrea Djibouti
    Quote Originally Posted by Shaikorth View Post
    Does Remedello also affect Mycenaean in Greeks? If it does overfitting should be a given?
    Quote Originally Posted by Gravetto-Danubian View Post
    For me, it barely affects things.
    Barely or doesn't at all affect things for me as well:

    [1] "distance%=0.0677 / distance=0.000677"

    Greek

    Mycenaean:I9041 39.55
    Mycenaean:I9006 18.95
    Mycenaean:I9010 14.50
    Sintashta:RISE395 11.35
    Corded_Ware_Germany:I0104 8.45
    Polish 3.65
    Corded_Ware_Germany:I1538 1.85
    Belarusian 1.65
    Yamnaya_Samara:I0443 0.05
    Mycenaean:I9033 0.00
    Remedello_BA:RISE486 0.00
    Remedello_BA:RISE487 0.00
    Remedello_BA:RISE489 0.00


    [1] "distance%=0.0776 / distance=0.000776"

    Greek

    Mycenaean:I9041 37.50
    Mycenaean:I9010 19.45
    Mycenaean:I9006 16.65
    Corded_Ware_Germany:I0104 10.85
    Polish 6.70
    Corded_Ware_Germany:I1544 4.45
    Yamnaya_Samara:I0444 1.85
    Corded_Ware_Germany:I1536 1.65
    Corded_Ware_Germany:I1538 0.90
    Mycenaean:I9033 0.00
    Remedello_BA:RISE486 0.00
    Remedello_BA:RISE487 0.00
    Remedello_BA:RISE489 0.00



    [1] "distance%=0.092 / distance=0.00092"

    Greek

    Mycenaean:I9041 43.45
    Mycenaean:I9006 15.40
    Mycenaean:I9010 11.55
    Polish 10.55
    Yamnaya_Samara:I0444 9.55
    Remedello_BA:RISE489 5.65
    Yamnaya_Samara:I0443 3.85
    Mycenaean:I9033 0.00
    Remedello_BA:RISE486 0.00
    Remedello_BA:RISE487 0.00


    So it seems the reason Tuscans show a good amount of Remedello in comparison is, well, because they're Italian.

  10. The Following 3 Users Say Thank You to Awale For This Useful Post:

     Agamemnon (08-08-2017), Hando (09-01-2017), Power77 (08-08-2017)

  11. #576
    Registered Users
    Posts
    69
    Sex
    Y-DNA
    J2a-M319

    Quote Originally Posted by Awale View Post
    Barely or doesn't at all affect things for me as well:

    [1] "distance%=0.0677 / distance=0.000677"

    Greek

    Mycenaean:I9041 39.55
    Mycenaean:I9006 18.95
    Mycenaean:I9010 14.50
    Sintashta:RISE395 11.35
    Corded_Ware_Germany:I0104 8.45
    Polish 3.65
    Corded_Ware_Germany:I1538 1.85
    Belarusian 1.65
    Yamnaya_Samara:I0443 0.05
    Mycenaean:I9033 0.00
    Remedello_BA:RISE486 0.00
    Remedello_BA:RISE487 0.00
    Remedello_BA:RISE489 0.00


    [1] "distance%=0.0776 / distance=0.000776"

    Greek

    Mycenaean:I9041 37.50
    Mycenaean:I9010 19.45
    Mycenaean:I9006 16.65
    Corded_Ware_Germany:I0104 10.85
    Polish 6.70
    Corded_Ware_Germany:I1544 4.45
    Yamnaya_Samara:I0444 1.85
    Corded_Ware_Germany:I1536 1.65
    Corded_Ware_Germany:I1538 0.90
    Mycenaean:I9033 0.00
    Remedello_BA:RISE486 0.00
    Remedello_BA:RISE487 0.00
    Remedello_BA:RISE489 0.00



    [1] "distance%=0.092 / distance=0.00092"

    Greek

    Mycenaean:I9041 43.45
    Mycenaean:I9006 15.40
    Mycenaean:I9010 11.55
    Polish 10.55
    Yamnaya_Samara:I0444 9.55
    Remedello_BA:RISE489 5.65
    Yamnaya_Samara:I0443 3.85
    Mycenaean:I9033 0.00
    Remedello_BA:RISE486 0.00
    Remedello_BA:RISE487 0.00


    So it seems the reason Tuscans show a good amount of Remedello in comparison is, well, because they're Italian.
    Could you explain what nmonte is exactly, is it like a gedmatch oracle?

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

     Hando (09-01-2017)

  13. #577
    Gold Member Class
    Posts
    547
    Sex
    Location
    Gulf of Fars
    Ethnicity
    Somali
    Y-DNA
    E-V32
    mtDNA
    N1a

    Somalia Ethiopia Eritrea Djibouti
    Quote Originally Posted by Sorcelow View Post
    Could you explain what nmonte is exactly, is it like a gedmatch oracle?
    It runs a Monte Carlo Simulation where, to put it simply, you basically give it a "Target" population (in this case, the target was Greeks) and then you give it any number of "Input" populations (in the above cases, the inputs were basically groups like the Mycenaeans, Poles, Corded-Ware, Remedello and so on) and and it then basically tries to create mixture models between the input populations that come as close as possible to replicating the profile of the target population based on something like where that population clusters on a PCA. GD and I are using the Global 10 PCA positions of all these populations in this case. As per the readme file itself:

    Technically the program performs a Monte Carlo simulation, evolutionary guided by the computed distances.

    The absolute best fit is 0 where the target and the model are essentially identical but it's good to wonder about over-fitting and be skeptical if if it's giving you really low values like 0.1something or 0.06, try to make sure the model actually makes sense too and fits with other data in this case. You've entered bad fit territory once the distance exceeds "1" and real bad once it's like "2" or above. As an example here's what Somalis get when you throw in singular input populations such as Somali, Somali_Kenya, Oromos and Tirginyas:

     
    [1] "distance%=0 / distance=0"

    Somali

    Somali 100

    [1] "distance%=0.4608 / distance=0.004608"

    Somali

    Somali_Kenya 100

    These Kenyan samples are mostly identical to the normal Somali sets out there but they have a few somewhat Oromo-like admixed individuals among them which is what creates the disconnect but 0.4 is still pretty close, though not identical, obviously.

    [1] "distance%=0.705 / distance=0.00705"

    Somali

    Ethiopian_Oromo 100

    Considering you could model these Oromos as being close to a 70% or so Somali-like population- :

    [1] "distance%=0.298 / distance=0.00298"

    Ethiopian_Oromo

    Somali 78.2
    Mota 14.2
    BedouinB 7.6

    [1] "distance%=0.3002 / distance=0.003002"

    Ethiopian_Oromo

    Somali 73.80
    Ethiopian_Ari_blacksmith 20.25
    Yemenite_Jew 5.95

    - Not a bad fit at all but I tend to be more impressed once something goes below 0.5.

    [1] "distance%=1.6731 / distance=0.016731"

    Somali

    Ethiopian_Tigray 100

    To get how "close" Somalis and Tigrinyas are to some extent you can go here and here or even, if it's more your speed, look at my Oracle single pop sharing results:

    # Population (source) Distance
    1 Somali 3.03
    2 Ethiopian_Oromo 7.13
    3 Ethiopian_Wolayta 8.81
    4 Ethiopian_Amhara 13.96
    5 Ethiopian_Tigray 16


    It's run with R and you can learn how to use it by going here. Hope that helps.

  14. The Following 9 Users Say Thank You to Awale For This Useful Post:

     Agamemnon (08-08-2017), alexfritz (08-08-2017), Asimakidis (08-08-2017), ffoucart (08-08-2017), Hando (09-01-2017), palamede (08-08-2017), Power77 (08-08-2017), Sorcelow (08-08-2017), Targum (08-08-2017)

  15. #578
    Registered Users
    Posts
    69
    Sex
    Y-DNA
    J2a-M319

    Quote Originally Posted by Awale View Post
    It runs a Monte Carlo Simulation where, to put it simply, you basically give it a "Target" population (in this case, the target was Greeks) and then you give it any number of "Input" populations (in the above cases, the inputs were basically groups like the Mycenaeans, Poles, Corded-Ware, Remedello and so on) and and it then basically tries to create mixture models between the input populations that come as close as possible to replicating the profile of the target population based on something like where that population clusters on a PCA. GD and I are using the Global 10 PCA positions of all these populations in this case. As per the readme file itself:

    Technically the program performs a Monte Carlo simulation, evolutionary guided by the computed distances.

    The absolute best fit is 0 where the target and the model are essentially identical but it's good to wonder about over-fitting and be skeptical if if it's giving you really low values like 0.1something or 0.06, try to make sure the model actually makes sense too and fits with other data in this case. You've entered bad fit territory once the distance exceeds "1" and real bad once it's like "2" or above. As an example here's what Somalis get when you throw in singular input populations such as Somali, Somali_Kenya, Oromos and Tirginyas:

     
    [1] "distance%=0 / distance=0"

    Somali

    Somali 100

    [1] "distance%=0.4608 / distance=0.004608"

    Somali

    Somali_Kenya 100

    These Kenyan samples are mostly identical to the normal Somali sets out there but they have a few somewhat Oromo-like admixed individuals among them which is what creates the disconnect but 0.4 is still pretty close, though not identical, obviously.

    [1] "distance%=0.705 / distance=0.00705"

    Somali

    Ethiopian_Oromo 100

    Considering you could model these Oromos as being close to a 70% or so Somali-like population- :

    [1] "distance%=0.298 / distance=0.00298"

    Ethiopian_Oromo

    Somali 78.2
    Mota 14.2
    BedouinB 7.6

    [1] "distance%=0.3002 / distance=0.003002"

    Ethiopian_Oromo

    Somali 73.80
    Ethiopian_Ari_blacksmith 20.25
    Yemenite_Jew 5.95

    - Not a bad fit at all but I tend to be more impressed once something goes below 0.5.

    [1] "distance%=1.6731 / distance=0.016731"

    Somali

    Ethiopian_Tigray 100

    To get how "close" Somalis and Tigrinyas are to some extent you can go here and here or even, if it's more your speed, look at my Oracle single pop sharing results:

    # Population (source) Distance
    1 Somali 3.03
    2 Ethiopian_Oromo 7.13
    3 Ethiopian_Wolayta 8.81
    4 Ethiopian_Amhara 13.96
    5 Ethiopian_Tigray 16


    It's run with R and you can learn how to use it by going here. Hope that helps.

    Thank you very much for the explanation Awale, it seems that by these simulations modern Greeks come up as 3/4 Mycenaean and 1/4 some type of Slavic population. Its also really interesting to see individuals from Pontus derive nearly 1/3 of their ancestry from a Mycenaean like source. I dont have access to the global 10 calculator, but would it be possible for you to run by results based off another calculator? Thanks once again.

  16. The Following 6 Users Say Thank You to Sorcelow For This Useful Post:

     Agamemnon (08-08-2017), Asimakidis (08-08-2017), Coldmountains (08-08-2017), ffoucart (08-08-2017), Hando (09-01-2017), Power77 (08-08-2017)

  17. #579
    Registered Users
    Posts
    1,552
    Y-DNA
    I-S336
    mtDNA
    K1c1

    Quote Originally Posted by Awale View Post
    Barely or doesn't at all affect things for me as well:


    So it seems the reason Tuscans show a good amount of Remedello in comparison is, well, because they're Italian.
    He gets just 5% Remedello for Tuscans though. Is it dependant on individuals?

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

     Awale (08-09-2017), Hando (09-01-2017), Power77 (08-09-2017)

  19. #580
    Gold Class Member
    Posts
    7,300
    Sex
    Location
    United Kingdom
    Nationality
    British
    mtDNA
    H

    United Kingdom
    Quote Originally Posted by 123Vlodor123 View Post
    Vinca culture and Plocnik culture both from Neolithic Serbia created first copper smelting in Europe and the world.
    Yes I have published about copper smelting in Serbia (see below). What I am trying to get across is that archaeologists use certain terms for periods in prehistory in accordance with what people at the time used to make tools. In northern Europe archaeologists go direct from Neolithic (New Stone Age) to Bronze Age, as the Copper Age there was too short to be noticed. In southern Europe archaeologists use the term 'Copper Age' for the period in which that metal was smelted, but before the invention of Bronze. The period varies from country to country in dates, but in the Balkans it began early. The copper smelting at Belovode marks the start of the Copper Age in Europe. It is not the Stone Age any more.

    Extract from my book Ancestral Journeys:

    The Copper Age

    Even before smelting was invented, people were attracted by the colour and shine of natural copper. From about 10,000 BC copper was worked cold into beads and ornaments in the heartland of the Neolithic where Anatolia meets the Levant. By 8000 BC some within that core area had discovered that heat (annealing) made copper-working easier. Then around 5000 BC smelting and cast-copper objects appeared both east and west of the heartland: at Tal-i Iblis in Iran and Belovode in Serbia. This simultaneous surge of the same technology makes a single locus of invention likely, probably in eastern Turkey, the centre of the range of early smelting.

    Copper Age Europe

    The first metal to be worked anywhere in Europe was copper. The earliest appearance of the new technology was in the Balkans. Farmers had prospered on the rich, silt soils of the lower Danube basin. Hamlets in what is now Bulgaria, Romania and Serbia grew into solidly-built villages of multi-roomed houses. Rebuilt again and again on the same site, the remains of such villages form a mound or 'tell'. Pottery kilns fired at high temperatures paved the way for metallurgy. Smelted copper tools and ornaments began to circulate around 5000 BC. Gumelniţa is one such site in Romania which has lent its name to a wider culture stretching from the Danube estuary to Thrace. To the north of the Gumelniţa culture was the impressive Cucuteni-Tripolye culture, within present-day Romania, Moldova and Ukraine (see also p. 98). It emerged between the Carpathian Mountains and the Middle Dniester around 4800 BC and spread gradually northeast. To the west, in present-day Serbia, the long-lived Vinča settlement stood on the banks of the Danube from around 5500 to 4000 BC. It too gave its name to a wider Balkan culture ..
    .
    The Balkans had deposits of copper, which would be an attraction to metal-workers. They also had gold. Gold is too soft to use for tools, but it is the king of metals for personal adornment. It is easily worked, does not tarnish and gleams like the sun. The earliest gold objects in the world have been found in the Balkans. The wealth of the Varna necropolis (4600-4200 BC) in Bulgaria is astonishing. The bulk of the gold there was found in just three of the hundreds of graves. These three were also distinctive in having staffs or sceptres among the grave goods: symbols of royal or spiritual power. Social stratification had entered Europe. Today we take rank for granted. Leaders have been a part of European life for the whole of recorded history. Yet European prehistory shows little sign of marked distinctions between individuals until the Copper Age. Gold in burials becomes one of the strongest clues to high status. The golden crown was to become the emblem of royalty.

    The recent discovery of bronze in the Balkans dating to c. 4500 BC revises the history of that useful metal, harder than pure copper. Unlike later bronze, made from copper alloyed with tin, these early bronze items were smelted from naturally occurring tin-bearing copper ores. The result was a bright yellow colour which would mimic gold.
    Last edited by Jean M; 08-08-2017 at 10:51 PM.

  20. The Following 10 Users Say Thank You to Jean M For This Useful Post:

     Agamemnon (08-09-2017), ArmandoR1b (08-08-2017), Awale (08-09-2017), Cascio (08-08-2017), Coldmountains (08-08-2017), Hando (09-01-2017), Michał (08-08-2017), palamede (08-09-2017), Power77 (08-08-2017), spruithean (08-09-2017)

Page 58 of 74 FirstFirst ... 848565758596068 ... LastLast

Similar Threads

  1. Replies: 327
    Last Post: 08-06-2017, 01:56 PM
  2. Genetic Origins of Terriers - Any thoughts?
    By JohnHowellsTyrfro in forum Fauna
    Replies: 8
    Last Post: 06-11-2017, 07:10 AM
  3. [Split] The Jatts & Their Genetic Origins
    By speedyran in forum Autosomal (auDNA)
    Replies: 172
    Last Post: 07-25-2015, 07:09 PM
  4. Replies: 0
    Last Post: 02-04-2014, 05:07 AM
  5. Minoans were European
    By Grossvater in forum General
    Replies: 23
    Last Post: 10-11-2013, 05:34 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
  •