Page 3 of 4 FirstFirst 1234 LastLast
Results 21 to 30 of 32

Thread: Does it matter where you get your raw data from?

  1. #21
    Registered Users
    Posts
    2,806
    Sex
    Location
    America
    Ethnicity
    North & Ionian Seas
    Nationality
    American
    Y-DNA (P)
    I1 (P109)

    England Italy Germany Scotland
    Quote Originally Posted by Censored View Post
    That's how it should be but apparently its not. The segments they analyze are different for each company so the raw data will be different.
    wait, this can't be so.

    your DNA is your DNA. while each company reviews only a portion of it and different than the next company.... the raw data does not change. for easy numbers..... if your raw data is 1-20 and company A only reviews 1-10 for a prediction, the raw data is still 1-20.

  2. #22
    Banned
    Posts
    2,951
    Sex
    Location
    America
    Ethnicity
    Telgu/marathi

    Siberian Tatars Tajikistan
    Quote Originally Posted by JerryS. View Post
    wait, this can't be so.

    your DNA is your DNA. while each company reviews only a portion of it and different than the next company.... the raw data does not change. for easy numbers..... if your raw data is 1-20 and company A only reviews 1-10 for a prediction, the raw data is still 1-20.
    No, I don't think that's how it works. The raw data itself is only a small portion of your DNA sequences, and it varies by company. So it matters which company gets the saliva sample.

  3. The Following User Says Thank You to Censored For This Useful Post:

     JerryS. (02-25-2018)

  4. #23
    Registered Users
    Posts
    2,806
    Sex
    Location
    America
    Ethnicity
    North & Ionian Seas
    Nationality
    American
    Y-DNA (P)
    I1 (P109)

    England Italy Germany Scotland
    Quote Originally Posted by Censored View Post
    No, I don't think that's how it works. The raw data itself is only a small portion of your DNA sequences, and it varies by company. So it matters which company gets the saliva sample.

    are you sure? where did you get that information? I'm not saying you made it up, I just never heard that before or at least not the way you put it.

    thanks.

  5. #24
    Banned
    Posts
    2,951
    Sex
    Location
    America
    Ethnicity
    Telgu/marathi

    Siberian Tatars Tajikistan
    Quote Originally Posted by JerryS. View Post
    are you sure? where did you get that information? I'm not saying you made it up, I just never heard that before or at least not the way you put it.

    thanks.
    Yes. http://www.geneticlifehacks.com/ance...aw-data-files/

    They can test a different number of nuceotides and they don't match on some of them. That results in your data differing between different raw data sets, although I'm not sure by how much.

  6. The Following User Says Thank You to Censored For This Useful Post:

     JerryS. (02-25-2018)

  7. #25
    Registered Users
    Posts
    26
    Sex
    Location
    New Mexico
    Ethnicity
    European
    Nationality
    Brazil, Italy
    mtDNA (M)
    H3g

    Brazil Italy Italy Veneto Spain Portugal Switzerland
    I tested on 23andMe last August, and AncestryDNA on October. Both results registered on Genesis GEDMatch. Although I am my strongest match on each one, the list of matches for each company is different. I tried to located some of the top ones in the other list but no success. I asked in another forum why did this happened and the answer was "this is impossible, you didn't read it right"... I wish people would answer only if they actually know what it is being talked about.
    Italian, Portuguese, Spanish, Swiss, all mixed together to make a real Brazilian. Trying to find the history of my Spanish side for decades, now on DNA as a last resource. I think my grandfather's family changed names when they immigrated. Hidden Content I accept help. Anyone from Málaga over here? Hidden Content Tested on 23andMe, v.5 chip not compatible anywhere else yet.

  8. The Following User Says Thank You to Vicmalaga For This Useful Post:

     JerryS. (02-26-2018)

  9. #26
    Registered Users
    Posts
    1,566
    Sex
    Location
    Pittsburgh, PA, USA
    Ethnicity
    Ger.-Brit.-Catalan-more
    Nationality
    (U.S.) American
    Y-DNA (P)
    R-YP619*
    mtDNA (M)
    H1bg

    United Kingdom Germany Bayern Catalonia France Ireland Switzerland
    I have personally taken 23andMe's v2 and v3 tests; FTDNA's Family Finder; and Ancestry DNA v1 and v2. None of these tests exactly the same SNPs as any other, though some of the tests overlap more than others.

    As a specific example of differences between two different chip versions just from the same company, let's consider my father's and my eldest brother's files from 23andMe.

    My father's file is a combination v2/v2 file, and my brother's is v4. My father's file, copied into Excel, consists of 991,792 rows of data. Each of these rows includes a different SNP from from my father's autosomes 1-22, his X chromosome, his Y chromosome, and his mtDNA.

    By contrast, my brother's file copied into Excel consists only of 610,545 rows of data. But it isn't simply a matter of my brother's file having over 300 thousand fewer SNPs that my father's file. In fact, my brother's file includes many hundreds -- likely thousands -- of SNPs that my father's does not.

    I also have verified for myself that my Ancestry, FTDNA, and 23andMe files differ in more than just size. Each tests for some SNPs that one or both of the other two do not.
    Besides British-German-Catalan, ancestry includes smaller amounts of French, Irish, Swiss, Choctaw & another NA tribe, possibly Catawba. Avatar picture is: my father, his father, & his father's father; baby is my eldest brother.

    GB

  10. The Following User Says Thank You to geebee For This Useful Post:

     Robert1 (03-01-2018)

  11. #27
    Registered Users
    Posts
    1,566
    Sex
    Location
    Pittsburgh, PA, USA
    Ethnicity
    Ger.-Brit.-Catalan-more
    Nationality
    (U.S.) American
    Y-DNA (P)
    R-YP619*
    mtDNA (M)
    H1bg

    United Kingdom Germany Bayern Catalonia France Ireland Switzerland
    I used this tool --


    -- to compare my 23andMe v2/v3 file and my Ancestry v2 file. Here's the bottom line:

    442582 SNPs appear in both files

    0.142 % (629) of these 442582 common SNPs are NoCalls (i.e. NoCall in at least one file)

    0.026 % (113) of the 442582 SNP results differ

    83 differing SNPs are both homozygous
    5 differing SNPs are homozygous in File1 but not in File2
    22 differing SNPs are homozygous in File2 but not in File1

    Note that even when the two companies tested the same SNPs, they occasionally got different results. Obviously, this suggests that the call made by at least one of the company's was wrong. However, this was very rare -- less than 3 hundredths of a percent of the tested SNPs showed different results.

    There were somewhat more instances in which one or the other of the two companies had a "no call" at a particular SNP, but this was still pretty rare -- less than 2 tenths of a percent.
    Last edited by geebee; 02-26-2018 at 07:49 PM.
    Besides British-German-Catalan, ancestry includes smaller amounts of French, Irish, Swiss, Choctaw & another NA tribe, possibly Catawba. Avatar picture is: my father, his father, & his father's father; baby is my eldest brother.

    GB

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

     Robert1 (03-01-2018)

  13. #28
    Registered Users
    Posts
    2,806
    Sex
    Location
    America
    Ethnicity
    North & Ionian Seas
    Nationality
    American
    Y-DNA (P)
    I1 (P109)

    England Italy Germany Scotland
    I'm wondering, if they would have evaluated all of your DNA would your ethnicities groups change, or would the percentages change? I think they would because when I use the Admixture Proportions by Chromosome on GEDmatch some of my ethnicities show considerable presence on some chromosomes and lightly on some while completely absent on others.

  14. #29
    Banned
    Posts
    2,951
    Sex
    Location
    America
    Ethnicity
    Telgu/marathi

    Siberian Tatars Tajikistan
    Quote Originally Posted by JerryS. View Post
    I'm wondering, if they would have evaluated all of your DNA would your ethnicities groups change, or would the percentages change? I think they would because when I use the Admixture Proportions by Chromosome on GEDmatch some of my ethnicities show considerable presence on some chromosomes and lightly on some while completely absent on others.
    I want to believe their current method is accurate but unfortunately I don’t think it is. Just the massive visitations between different companies in assigned ancestries based on the same raw data alone is proof enough of how much fluctuation there can be.

  15. The Following User Says Thank You to Censored For This Useful Post:

     JerryS. (02-27-2018)

  16. #30
    Registered Users
    Posts
    1,566
    Sex
    Location
    Pittsburgh, PA, USA
    Ethnicity
    Ger.-Brit.-Catalan-more
    Nationality
    (U.S.) American
    Y-DNA (P)
    R-YP619*
    mtDNA (M)
    H1bg

    United Kingdom Germany Bayern Catalonia France Ireland Switzerland
    Quote Originally Posted by Censored View Post
    I want to believe their current method is accurate but unfortunately I don’t think it is. Just the massive visitations between different companies in assigned ancestries based on the same raw data alone is proof enough of how much fluctuation there can be.
    But it isn't always "based on the same raw data", nor is the same algorithm necessarily used in all cases. That's the point. Some SNPs are more "ancestry informative" than others.

    Different companies also do not all use the same reference panels. Yet two companies may use nearly the same name for a category, even when the categories don't necessarily mean exactly the same thing. For example, one company may use "British & Irish" while another uses "Great Britain". These may sound similar, but how similar they actually are is going to depend on exactly what the composition of each reference group is.

    For example, if "Irish" is excluded from the "Great Britain" group, then how can you really compare it to a group that includes British and Irish? And when you say, "British", presumably that should include English, Scottish, and Welsh. But is this actually true of the reference groups?

    And how large are these reference groups? How reliable is a reference group that is made up of only 8 members, for example? Or a reference group drawn from just a small part of the region is supposedly represents?

    The variation in different company's results isn't solely the product of what SNPs or reference groups are used, or what software is used. But since each of these is different for different companies, it's hard to see why you would expect exactly the same results. This is especially true if categories are poorly designed. What, precisely, is meant by "Europe West"? How can you compare a "Europe West" result at one company to another company that also uses this term -- if you find one -- if you don't know that they mean exactly the same thing at both companies? Or what if, as I'd suggested previously, the terms are merely similar -- not identical?

    So when you say, "I want to believe their current method is accurate", one problem is that there is not a single current method. Different companies evaluate different SNPs, use different ancestral categories, different reference panels, and different algorithms. So of course they have different results.
    Besides British-German-Catalan, ancestry includes smaller amounts of French, Irish, Swiss, Choctaw & another NA tribe, possibly Catawba. Avatar picture is: my father, his father, & his father's father; baby is my eldest brother.

    GB

Page 3 of 4 FirstFirst 1234 LastLast

Similar Threads

  1. Family finder, does your maternal line matter?
    By utR! in forum Autosomal (auDNA)
    Replies: 7
    Last Post: 09-15-2017, 06:57 AM
  2. New map of Universe's dark matter
    By MikeWhalen in forum Natural Sciences
    Replies: 0
    Last Post: 08-03-2017, 08:40 PM
  3. New theory of gravity might explain dark matter
    By Amerijoe in forum Astronomy
    Replies: 0
    Last Post: 11-08-2016, 10:21 PM
  4. For groups in conflict, genes matter
    By Amerijoe in forum Human Evolution
    Replies: 5
    Last Post: 05-10-2016, 07:52 AM
  5. For groups in conflict, genes matter
    By rock hunter in forum General Sociology/Ethnology
    Replies: 0
    Last Post: 05-08-2016, 12:55 PM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •