Tuesday, May 01, 2018

DNA matches & segment analysis

I checked my One-To-Many match listing at GEDmatch for the first time in about a week and found 2 new (above 30 cM) matches.   One 41.6 cM and the other 35.8 cM.  My normal practice is to click on the L of the match to take a look at their match list and use the CTRL + F function to find my email addy on the match list.  That gives me an idea of which of the tests other than mine this person matches.

The first kitt that I admin which is listed on their match list(abt 65 lines down) is my own kit followed shortly by my brother's test.   At this point, I'm thinking that the match is likely on Dad's side as my Mom tested and I haven't seen her name on the page.   After looking down 5 more lines I see another kit I admin.  This is a match that neither I nor my brother has any relation --my daughter's Paternal 2nd Great Aunt(a Mosley/Hogland connection?).  Clicking to the next found occurrence of my email addy(which is about one click of the Page Down button away)  I see that this person also shares DNA with my daughter's paternal Great Aunt(niece of previously mentioned 2ndGr Aunt). On to the next match for a kit I admin which is my Mom's kit,  followed shortly by my daughter's kit.   At this point, I'm thinking the match is on Mom's side.  But wait, there's more.   The last match is my Paternal Aunt.

I can't think of a better example to illustrate the importance of a chromosome browser to establish whether a match is Paternal, Maternal or in this case both.  Segment analysis is extremely important.

While both of my parents have early Tennessee ancestors, I've not found any In-Common ancestors and I have most of the lines back to 4th Greats.  Given that they are both heavily colonial there could likely be some connections once we are back in the 1600s.  Occasionally, I do find cousins that I have in common with both sides which is what is happening in this case.

From this new match's list, I selected each kit adminned by me, selected Visualization Options and then Chromosome Browser(2D Chr Browser).

Once I view the matches of this person to my adminned kits in the 2D Chromosome browser, I can see what is happening.  The image below shows the segments and is a table format I made using the information in the chromosome browser.





The Chromosome 5 segment is one that my brother and I got from our Mom.  I passed that segment on to my daughter.  The Chromosome 12 segment is one that my daughter didn't get but is somewhere down her Paternal Mosley/Hogland lines.  The Chromosome 17 segment is one that my brother and I received from our Dad(never tested)

What's my takeaway from this?

 I can never assume sides from the One To Many lists.  Dad didn't test and there is nothing to compare from his side unless by chance they also match his sister.

While segment analysis is a must,  it's not something we will ever be able to do at Ancestry in my opinion. This match tested at Ancestry which I can tell by the A at the beginning of his kit number.  .I haven't been able to find him in my matches at Ancestry, but at least I have info for the segments.  AncestryDNA match list could really use a reliable search function but that is something they need to work on site wide.  I am not sure if GEDmatch Genesis will allow for identification of where the Match tested.  There was a column for it in the early GEDmatch Genesis Beta but there doesn't appear to be a way to tell with the current display.  I guess we will know once they have combined the two GEDmatch databases

A quick check of my Visual Phasing of myself and my two siblings shows that the Chr 5 segment that my brother and I share is from our MGF and that our sister has a segment from our MGM at that portion of the Chr 5.  I've not finished my Chr 17 Visual Phasing so I can't compare that one yet.

So now I'm off to DNApainter to paint that segment of Chr. 5 as Maternal and the segment of Chr 17 as Paternal.  With any luck, they will overlap with segments of known cousins which I've already painted and perhaps I can narrow down the connection further.

No comments:

Post a Comment