tag:blogger.com,1999:blog-6533996127304587865.post8420561648076804576..comments2023-10-16T21:11:28.700+03:00Comments on Dodecad Ancestry Project: Comparing different ADMIXTURE runs using ZombiesDienekeshttp://www.blogger.com/profile/02082684850093948970noreply@blogger.comBlogger10125tag:blogger.com,1999:blog-6533996127304587865.post-78590071287078069632011-10-22T07:42:18.026+03:002011-10-22T07:42:18.026+03:00"Once Dodecad v4 is off the ground, and if I ..."<i><b>Once Dodecad v4</b> is off the ground, and if I find time to fully automate the process, I will perhaps try to map all my past calculators (i.e., the initial K=10, Dodecad v3, 'bat', 'euro7', 'weac', 'africa9') onto the new golden standard of the Project.</i>"<br /><br />Nice to hear something new is in the works. Looking forward to see what you have in store for us :-).<br /><br />Dienekes, I have just one request. Instead of accounting for ancestries in the <i>Information about project samples</i> thread; or at least in addition to that, could you a create a [freely editable] Google spreadsheet with columns for the Dodecad ID (DODnnnn); Ethnicity/Nationality; Y-DNA and mtDNA Haplogroups like the way <a href="https://docs.google.com/spreadsheet/ccc?key=0AuW3R0Ys-P4HdGE4eDh6emt1dUs2U2pXTkVjS0lsV1E&hl=en#gid=0" rel="nofollow">Zack from the Harappa Ancestry Project</a> has done? That would be very, very useful. Going through the numerous posts in the ancestry thread in order to look up a participant of interest's ethnicity is rather harrowing.Vasishtahttps://www.blogger.com/profile/16475698920004634252noreply@blogger.comtag:blogger.com,1999:blog-6533996127304587865.post-71921282695271256672011-10-21T08:53:43.480+03:002011-10-21T08:53:43.480+03:00You'll have to check with the MDLP to see how ...You'll have to check with the MDLP to see how these components were inferred.Dienekeshttps://www.blogger.com/profile/02082684850093948970noreply@blogger.comtag:blogger.com,1999:blog-6533996127304587865.post-55396039448257788962011-10-21T03:27:10.625+03:002011-10-21T03:27:10.625+03:00If it isn't too much trouble could you list th...If it isn't too much trouble could you list the populations that make up the Scandinavian, Western European and Celto-Germanic groups?Average Joehttps://www.blogger.com/profile/12203996329459638052noreply@blogger.comtag:blogger.com,1999:blog-6533996127304587865.post-50742848021285826352011-10-20T20:45:50.070+03:002011-10-20T20:45:50.070+03:00You should post an addendum with a comparison betw...<i>You should post an addendum with a comparison between Dodecad V3 and the latest Eurogenes K=10. </i><br /><br />a very good ideaOnur Dincerhttps://www.blogger.com/profile/05041378853428912894noreply@blogger.comtag:blogger.com,1999:blog-6533996127304587865.post-37450180897218297432011-10-20T19:39:37.101+03:002011-10-20T19:39:37.101+03:00If you keep doing it, and you don't figure out...<i>If you keep doing it, and you don't figure out a way to avoid major errors, like with the Ukrainians and Mordvinians, you'll give me a lot to blog about. It's gonna be fun.</i><br /><br />You have your concepts mixed. The results of Ukrainians and Mordvins are not attributable to the use of zombies. Approve my comment and you will learn something new!Dienekeshttps://www.blogger.com/profile/02082684850093948970noreply@blogger.comtag:blogger.com,1999:blog-6533996127304587865.post-29325803886935542842011-10-20T19:34:41.231+03:002011-10-20T19:34:41.231+03:00@Davidski, kindly approve my comment at your blog....@Davidski, kindly approve my comment at your blog.<br /><br />As for the zombie concept, it works just fine. I have never seen a major discrepancy between results generated using supervised ADMIXTURE+Zombies vs. those generated using DIYDodecad+allele frequencies.Dienekeshttps://www.blogger.com/profile/02082684850093948970noreply@blogger.comtag:blogger.com,1999:blog-6533996127304587865.post-53034518252670180852011-10-20T19:33:02.012+03:002011-10-20T19:33:02.012+03:00Whoops, truncated...
As I was saying, the zombie ...Whoops, truncated...<br /><br />As I was saying, the zombie supervised thing doesn't work in many cases. It was an interesting idea, but it only works for highly differentiated clusters.<br /><br />If you keep doing it, and you don't figure out a way to avoid major errors, like with the Ukrainians and Mordvinians, you'll give me a lot to blog about. It's gonna be fun.Davidskihttps://www.blogger.com/profile/04637918905430604850noreply@blogger.comtag:blogger.com,1999:blog-6533996127304587865.post-38332297369726234072011-10-20T19:21:19.971+03:002011-10-20T19:21:19.971+03:00The zombie concept doesn't work for intra-Euro...The zombie concept doesn't work for intra-European, or even intra-West Eurasian clusters.<br /><br />It's a mess all round for such closely relatedDavidskihttps://www.blogger.com/profile/04637918905430604850noreply@blogger.comtag:blogger.com,1999:blog-6533996127304587865.post-18402329206814743422011-10-20T19:18:09.408+03:002011-10-20T19:18:09.408+03:00If I find some time to automate the process, I wil...If I find some time to automate the process, I will map all the previous Dodecad calculators and a few ones from other projects onto the Dodecad v4 platform. It is a bit tedious to do the analysis I did here step-by-step.Dienekeshttps://www.blogger.com/profile/02082684850093948970noreply@blogger.comtag:blogger.com,1999:blog-6533996127304587865.post-66449881435948271782011-10-20T18:01:59.374+03:002011-10-20T18:01:59.374+03:00You should post an addendum with a comparison betw...You should post an addendum with a comparison between Dodecad V3 and the latest Eurogenes K=10.Eduardo Pintohttps://www.blogger.com/profile/09199139277049366469noreply@blogger.com