Announcement

Collapse
No announcement yet.

Evaluating a Big Y match

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • rrtipton1
    replied
    Originally posted by Brunetmj View Post
    Which begs the question just what does Family Tree consider a match?
    I am by no means an expert on this topic, but it has been my understanding that your BigY Match list is based on your set of Known SNPs. Anybody that matches you with no more than four differences is considered to be a match. The match list is normally sorted by the number of Known SNP Differences, and the actual Non-Matching Known SNPs are shown in the next column.

    In my own case, I have 38 matches listed. Of these, 4 are exact matches, 11 have one mismatched SNP, 12 have 2 mismatched SNPs, 9 have 3 mismatches, and 2 have 4 mismatches. I happen to be in haplogroup I-L813. All but one of my matches are also in I-L813. The one exception is I-Z74, which is one step up the FTDNA haplotree. He does not have one of the SNPs that define the I-L813 haplogroup.

    All of the other Non-Matching Known SNPs that are listed appear to be private SNPs to either myself or my matches. For example six of my matches share the SNP known as F454. The other 34 matches are F454-. According to the ISOGG YBrowse utility, F454 was observed in haplogroup O3-M117. Not a reliable SNP from the looks of it, but it plays a significant part in the makeup of my match list.

    By the way, there is a pull down near the top of your match report labelled "Filter matches by subclade". If you pull that down, you will see a list of all of the haplogroups (according to the FTDNA haplotree) that appear in your match list. You can select one of those (if you have more than one) as a filter. Since I only have one match that is not in my own Hg, I usually don't bother.

    Leave a comment:


  • Armando
    replied
    Originally posted by dna View Post
    Until yesterday, I thought I knew the answer. Last night I was looking at matches of N and U, and realized that either FTDNA is inconsistent or there is something going on behind the scenes, as far as Big Y matching is concerned.

    Mr. W.

    P.S.
    FTDNA is inconsistent.
    For example, if there is a set of rules that governs what SNPs are being placed on SNP certificates, and which ones are not, it must be full of special cases and exceptions.
    It is because of the inconsistency of FTDNA that professional analysis by YFull, FGC, and Alex Williamson(if P312) are recommended.

    Leave a comment:


  • dna
    replied
    Originally posted by Brunetmj View Post
    Which begs the question just what does Family Tree consider a match?
    Until yesterday, I thought I knew the answer. Last night I was looking at matches of N and U, and realized that either FTDNA is inconsistent or there is something going on behind the scenes, as far as Big Y matching is concerned.

    Mr. W.

    P.S.
    FTDNA is inconsistent.
    For example, if there is a set of rules that governs what SNPs are being placed on SNP certificates, and which ones are not, it must be full of special cases and exceptions.

    Leave a comment:


  • Brunetmj
    replied
    The problem is that Big Y testers only see whether any of their matches shares novel variants with them.
    Which begs the question just what does Family Tree consider a match?

    Leave a comment:


  • T E Peterman
    replied
    We can't compare data that doesn't exist. The only way we can get data on whether people outside Big Y have the novel variant is for them to take the Big Y test.

    Timothy Peterman

    Leave a comment:


  • dna
    replied
    Originally posted by T E Peterman View Post
    There is a tab where all novel variants are listed. If you download the CSV file, you will get a list of all novel variant values & all known SNP values.

    You can then check off the novel variants against the match list where novel variants are only reported if they are shared with someone. I've done this with all 5 Big Y tests I manage.
    The problem is that Big Y testers only see whether any of their matches shares novel variants with them. There is no indication whether anybody else out of those who tested Big Y also has it.

    Let me go back to the grand uncle (U)/grand nephew (N) pair.

    U has two SNPs at positions 222xxxxx and 223xxxxx, N does not have them. I might have thought that these two (novel) SNPs are unique to U. But they are not! Although none of U matches has them, one of N matches has them! Only because I have access to both kits, I could see that...

    W. (Mr.)

    Leave a comment:


  • T E Peterman
    replied
    There is a tab where all novel variants are listed. If you download the CSV file, you will get a list of all novel variant values & all known SNP values.

    You can then check off the novel variants against the match list where novel variants are only reported if they are shared with someone. I've done this with all 5 Big Y tests I manage.

    Timothy Peterman

    Leave a comment:


  • dna
    replied
    @1798

    Originally posted by T E Peterman View Post
    They do post the novel variants. But they don't tell you which ones are unshared.

    Timothy Peterman
    Big Y testers only see whether any of their matches shares novel variants with them.

    At the minimum, FTDNA should expand the Big Y list of known SNPs by including thousands of SNPs discovered through Big Y (the SNPs named BYxxx and BYxxxx).

    W. (Mr.)

    Leave a comment:


  • T E Peterman
    replied
    They do post the novel variants. But they don't tell you which ones are unshared.

    Timothy Peterman

    Leave a comment:


  • 1798
    replied
    Originally posted by T E Peterman View Post
    The first thing I do with Big Y results (after requesting a BAN file) is to download the SNP list, which includes known SNPs & novel variants, into Excel. I delete the known SNPs, saving only the novel variants. I add a column called Shared With. Then, I go to the match list & look at the shared novel variant drop down list. I highlight each of them on the novel variant spreadsheet, making a record of the number of shared matches. After I'm done, I sort the Shared with column. I assign 0 to all of those unshared. I count the number of these unshared novel variants; they will, in the future, prove to be the most important piece of data coming from the Big Y test.

    Timothy Peterman
    It would be easier if FTDNA posted the novel variants of a tester's Big Y results. It would save the tester a lot of work. When I take the test it will be for the novel variants only that I will be interested in.

    Leave a comment:


  • K. L. Adams
    replied
    Originally posted by T E Peterman View Post
    The first thing I do with Big Y results (after requesting a BAN file) is to download the SNP list, which includes known SNPs & novel variants, into Excel. I delete the known SNPs, saving only the novel variants. I add a column called Shared With. Then, I go to the match list & look at the shared novel variant drop down list. I highlight each of them on the novel variant spreadsheet, making a record of the number of shared matches. After I'm done, I sort the Shared with column. I assign 0 to all of those unshared. I count the number of these unshared novel variants; they will, in the future, prove to be the most important piece of data coming from the Big Y test.

    Timothy Peterman
    Thank you for sharing your process. I'm going to give it a try when I receive my BAM file. Kevin

    Leave a comment:


  • T E Peterman
    replied
    The first thing I do with Big Y results (after requesting a BAN file) is to download the SNP list, which includes known SNPs & novel variants, into Excel. I delete the known SNPs, saving only the novel variants. I add a column called Shared With. Then, I go to the match list & look at the shared novel variant drop down list. I highlight each of them on the novel variant spreadsheet, making a record of the number of shared matches. After I'm done, I sort the Shared with column. I assign 0 to all of those unshared. I count the number of these unshared novel variants; they will, in the future, prove to be the most important piece of data coming from the Big Y test.

    Timothy Peterman

    Leave a comment:


  • Brunetmj
    replied
    What counts is unshared novel variants. A lot of the 78 are shared with everyone else in your haplogroup. To be clear, novel doesn't mean unique, it just means new, as in just discovered.
    For me that's an important statement . I was under the impression that novel variant meant unique to my family line. Instead it means as described in your statement . This means to evaluate a recent match a person needs to be in my unique subclade. I am alone in a sub branch of my subclade. So a person would have to posses these same snp's to be in a recent family line with the possibility of having some I do not have.

    Leave a comment:


  • T E Peterman
    replied
    Very interesting. I think 7 may be a bit too many as well.

    Would be nice to get some real data on this. If the 135 years, or 5 generation rule is correct, as I see it: there is a 25% chance that a novel variant occurred between my father (1918) & I, there is a 50% chance that a novel variant occurred between my grandfather (1892) & I, there is a 75% chance a novel variant occurred between my great-grandfather (1862) & I, there is a 100% chance a novel variant occurred between my great-great grandfather (1829) & I.

    To continue, there is a 100% chance one novel variant occurred between my great-great-great grandfather (1788) & I, and a 25% chance two novel variants occurred between my great-great-great grandfather & I.

    There is a 100% chance one novel variant occurred between my g-g-g-g grandfather (1755) & I, and a 50% chance two novel variants occurred between my g-g-g-g grandfather & I.

    Am I calculating this correctly?

    I think there is a 100% chance that one novel variant would occur between two men as you described: the MRCA is the father of one & the great-grandfather of the other.

    But the whole logic of 135 years per novel variant may be flawed. In some lines, several hundred years could pass before there is a novel variant. In other lines, several novel variants might be incurred with each generation. 135 is likely just an average.

    But, unlike with y-str data where the counts fluctuate, once a man incurs a novel variant, all of his patrilineal descendants would share that novel variant; the chances of a reverse mutation are rather small.

    Timothy Peterman

    Leave a comment:


  • dna
    replied
    Originally posted by 1798 View Post
    So 78x90 is around 7000 years for a TMRCA.
    No, please refer to my earlier post this is a grand uncle/grand nephew pair. That means only four (4) births are separating them.


    Originally posted by T E Peterman View Post
    What counts is unshared novel variants. A lot of the 78 are shared with everyone else in your haplogroup. To be clear, novel doesn't mean unique, it just means new, as in just discovered.

    I can see a benefit in testing people with closely related patrilines. Here is my thinking. Many of the early immigrants to America were born between 1580 & 1650. Assuming 135 years per novel variant, consider the years 1580, 1715, 1850, 1985. Such an immigrant may have descendants who have incurred as many as 3 novel variants since immigrating. If someone born in 1985 had no unshared novel variants with a match, we would know that the MRCA for the two was likely born after 1850. If he had one unshared novel variant, we would know that the MRCA was likely born after 1715. Two unshared novel variants: MRCA likely born after 1580. These tiny subclades would be a more sound way of determining distance to the MRCA than y-str data.

    Timothy Peterman
    This is a grand uncle/grand nephew pair (father of the first man is great grandfather of the second man). One has two (2) unshared novel variants, the other has five (5) unshared novel variants. In my opinion, it is unlikely that there are 7 genealogically relevant SNPs that separate them. FTDNA possibly can verify in their entire database, whether these are SNPs that occur at many locations in the haplotree. Maybe these private SNPs, but we are not investing into more Big Y tests in that particular family branch.

    W. (Mr.)

    Leave a comment:

Working...
X