Question about siblings auDNA tests.

+4 votes
73 views
I and three of my siblings have all taken DNA test.

Looking at my test, I have some DNA segments that are triangulated outside my family group. Some of these I can designate as maternal or paternal.

Looking at my siblings test I am wondering if I can make the following assumption.

When they triangulate at a location and I can identify their match as paternal AND they do not match me at the same segment AND I have a match at that same location which I can triangulate as paternal grandfather THEN I can conclude their match is paternal grandmother.
asked in Genealogy Help by Lance Martin G2G6 Mach 6 (66.2k points)

2 Answers

+1 vote
 
Best answer
See if this answers your question. With siblings, you are labeling Regions of DNA for each Sibling, Half-Identical Regions (HIR) or Full Identical Regions (FIR). The following is only to answer your question.

The first step is to take all your known ½ cousins and beyond relatives who have been auDNA tested and record the HIR, segment information,  in a spreadsheet.  Add two Columns Paternal and Maternal.

For each HIR, enter into only one of the columns, either Grandfather or Grandmother, leaving the other column blank.

For each combination of siblings, comparing each known DNA cousin, those segments that DO NOT TRIANGULATE, since there is no shared segment (no overlap), you will need to add a line into one of the siblings and enter the alternative path.  If Grandfather is one sibling, you will add grandmother into the other.

But there are things you could/should do to increase your confidence.

1. When you add a line, identify the line as predicted.

2. When comparing siblings, about 25% of the DNA do not share either paternal or maternal grandparent.  For those Regions, there should not be the same paternal or maternal grandparent.

3. Basically, look for inconsistencies.

Triangulation is used when you don’t know how you are related. So now add those unknown DNA Cousins with segments.

When you have 4 siblings, for most matches, you should not have to rely on predicted segments.

Generally speaking, we begin with the 4 possible paths via grandparents we are attempting to narrow down the path to 2 out of the 4.

You are attempting to create a decision tree based on a solid foundation.

The next step would be to identify within each of these 4 paths, the next level.  For each grandparent, what HIR is from the great-grandmother and what is from the great-grandfather.

There is no need to triangulate.

I would suggest that you try Lazarus.

For a single known ½ first cousin or beyond, enter that person into group b, and the 4 siblings into group A. experiment with the size of the segments to see if it really makes a difference.  You will see the segments that are used from each sibling to create the kit, and other information that may be educational.

In your original post, you said that you could identify the match as paternal, How, without also knowing if the match is via the paternal grandfather or grandmother.

Are you relying on yDNA?
answered by Ken Sargent G2G6 Mach 5 (56.5k points)
selected by Lance Martin
Ill have to read that a few times and see if I can get it to work. Thank you very much for that information.

I have a few very lucky DNA tests that I have arranged. Besides having a first cousin (our mom's were sisters and her dad's family was a recent immigrant from Belgium) which give me my maternal matches, I have a half-first cousin on my fathers side, a half second cousin on my mothers side, and a half third cousin. That plus a few other full cousins and I think I should have enough data to find some segments which I share with through my GG Grandmother who is my brick wall, and her son in law which I think is perhaps related to the Swafford's of Tennnessee in some manner.

I have noticed that I do have a lot of crossover matching in a paternal and maternal line between 1700-1800 as both were in Tennesse and the Carolinas, but other than that, most matches are pretty distinct.

When working with my 4 siblings, I was also thinking at some point, the DNA of certain ancestors is knocked out of the record. So, if can identifiy my grandfathers mother in my DNA at a certain point, I will never have his father at that point in any of my sibilings DNA. That means, at that point in the DNA in my siblings if they do not match me, they are either junk or belong to one of my other great grandparents.
What tool do you used to identify full identical and half identical?
Ah. I found the Full Identical Half Identical on GEDMATCH under one to one repair with graphic.

I usually don't view the graphics.
0 votes
Given the usual caveats about size of the matches being big enough to give reasonable confidence, then your logic looks spot on to me.
answered by Derrick Watson G2G6 Mach 1 (19.5k points)

Related questions

+19 votes
8 answers
+5 votes
0 answers
+17 votes
0 answers
+8 votes
3 answers
106 views asked Sep 8, 2017 in Policy and Style by Jim Parish G2G6 Pilot (125k points)
+6 votes
1 answer
141 views asked Dec 12, 2017 in The Tree House by Kerry Larson G2G6 Mach 7 (75.7k points)
+14 votes
5 answers
251 views asked Sep 7, 2016 in The Tree House by Peter Roberts G2G6 Pilot (421k points)
+10 votes
6 answers
371 views asked May 12, 2016 in Policy and Style by Ken Sargent G2G6 Mach 5 (56.5k points)
+1 vote
0 answers

WikiTree  ~  About  ~  Help Help  ~  Search Person Search  ~  Surname:

disclaimer - terms - copyright

...