Announcement

Collapse
No announcement yet.

Inferring Human Colonization History Using a Copying Model

Collapse
X
Collapse
 
  • Page of 1
  • Filter
  • Time
  • Show
Clear All
new posts
Previous template Next
  • #1

    Inferring Human Colonization History Using a Copying Model

    I would like to take this test. I wonder when it will be available for us on the market?
    Inferring Human Colonization History Using a Copying Model
    http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pgen.1000078
    Author SummaryHumans like to tell stories. Amongst the most captivating is the story of the global spread of modern humans from their original homeland in Africa. Traditionally this has been the preserve of anthropologists, but geneticists are starting to make an important contribution. However, genetic evidence is typically analyzed in the context of anthropological preconceptions. For genetics to provide an accurate and detailed history without reference to anthropology, methods are required that translate DNA sequence data into histories. We introduce a statistical method that has three virtues. First, it is based on a copying model that incorporates the block-by-block inheritance of DNA from one generation to the next. This allows it to capture the rich information provided by patterns of DNA sharing across the whole genome. Second, its parameter space includes an enormous number of possible colonization scenarios, meaning that inferences are correspondingly rich in detail. Third, the inferred colonization scenario is determined algorithmically. We have applied this method to data from 53 human populations and find that while the current consensus is broadly supported, some populations have surprising histories. This scenario can be viewed as a movie, making it transparent where statistical analysis ends and where interpretation begins.
    Tags: None
  • #2
    This test will probably never arrive as a consumer test, however if your smart learner you can design your own test using STRUCTURE, the Stanford 650k dataset for 52 populations and your deCODEme.com testdata.

    Comment

    • #3
      Originally posted by Noaide
      This test will probably never arrive as a consumer test, however if your smart learner you can design your own test using STRUCTURE, the Stanford 650k dataset for 52 populations and your deCODEme.com testdata.
      Noaide, you're smart. offer the test to others!

      Comment

      • #4
        Originally posted by Noaide
        This test will probably never arrive as a consumer test, however if your smart learner you can design your own test using STRUCTURE, the Stanford 650k dataset for 52 populations and your deCODEme.com testdata.
        How can I get hold of the Stanford 650K dataset?

        Comment

        • #5
          Originally posted by Hando
          How can I get hold of the Stanford 650K dataset?
          Its here http://shgc.stanford.edu/hgdp/files.html

          However it is useless for you unless you have tested your own dataset.

          Comment

          • #6
            Originally posted by tomcat
            Noaide, you're smart. offer the test to others!
            I would then probably vioalte both the STRUCTURE licence and missuse the HGDP panel free availability.

            Comment

            • #7
              Noaide,

              Too BAD!!

              Are the Stanford SNP's covered by DeCodeMe or 23andMe's assays? My software does not support the Stanford file format, what Native American populations are part of the Standford/HGDP database?

              Comment

              • #8
                Originally posted by tomcat
                Are the Stanford SNP's covered by DeCodeMe or 23andMe's assays? My software does not support the Stanford file format, what Native American populations are part of the Standford/HGDP database?
                I suspect deCODEme and 23andme use different standards, at least if you compare identical rs markers for medical traits analysed by both companies you will find that they use different standards for some of these.

                deCODEme's dataset do for sure use identical standard to the one used by Stanford's 650k file, most of the autosomal and x snp's used in the Stanford file is included in the deCODEme beadchip.

                There are 5 native american populations, the ones in yellow on this map are included here.

                There is nothing mysterious about the Stanford file, its a 2 gig text file zipped, to unpack it I recommend Allzip a freeware package availble everywhere on the net.

                Also an advantage if your male, you will find your x-snp very useful (because their phased) if you have native american ancestry in your upstream x-snp trail, possibly you could locate native american x-snp haploblocks, but this is for more advanced users.

                Comment

                Previous template Next
                Working...
                X