Your TGs are pretty unique!

A Triangulation Concept

I often get questions along the lines of: “do my Matches have the same TGs?” or “can I form TGs for my 11 kits in one spreadsheet!” The answers are an emphatic: “NO!” and “NO!” Most of us are pretty ego-centric with our DNA analysis – and this is good! And while we are the center of our own universe of DNA segments, each of our Matches is, likewise, the center of their universe of DNA segments. Each of us gets random segments of DNA from our ancestors – random size segment(s) and random placement somewhere on our chromosomes. Once we get past the 2C (2nd cousin) or 3C level, it is quite amazing that we share DNA with more distant cousins at all. But we have many, many cousins, and many of them beat the odds and we share a DNA segment.  However, this does NOT mean that we and a Match both got the exact same DNA segment from a Common Ancestor (CA) – that very rarely happens.  We get a segment, and our Match gets a segment, and what we “see” in a Chromosome Browser is the portion of our individual segments from the CA that overlaps – the shared part of our segments. When we form a TG with various Match-segments (most matching each other), there are usually fairly well defined start and end locations to the TG*. Of the shared segments in a TG: some will start at the start of the TG, and end before the end of the TG; some will end at the end of the TG; some will “float” within the TG; and some, particularly with closer cousins, will be larger than the TG. These are all normal and expected [see an example in Figure 6 here]. The point of this concept is that each of the Matches in the TG will have their own unique TG – representing the full segment they got from the CA. If your shared segment with a Match, aligns with the start location of a TG, there is a very good chance that the Match’s segment from the CA began before yours did, and the Match’s TG has an earlier start location. Try this experiment: Take two (or more) Matches at GEDmatch with shared segments that start at the start of one of your TGs, and compare them to each other. Often their overlapping segment will start before your TG. In fact, using the Shared Segment search utility at GEDmatch, you can probably find other Matches that match in a Match’s TG, that don’t appear in your TG (and those Matches have segments that don’t overlap enough with your DNA to form a shared segment). The bottom lines for this concept are that Triangulation should be done on one “base” person (usually you) at a time; and there is more to the DNA passed down from an Ancestor than what is shown by any one descendant’s TG. Each of our TGs are pretty unique, and our Matches will not have the same TGs or chromosome map.

[*Sometimes the TG start and end locations are a little fuzzy (see here), but our focus should be on the bulk of the TG.]

08A Segment-ology: Your TGs are pretty unique! Concept by Jim Bartlett 20170907

4 thoughts on “Your TGs are pretty unique!

  1. Given that this post is about the uniqueness of triangulation groups, I have a question about how best to relate another person’s TG to one’s own research efforts. I created a triangulation group using a 2nd cousin as the reference person. All of the matches in the TG overlap on the same chromosome. One end of the triangulation group is anchored by another known cousin, and the other end is anchored by my father. All of the candidates in the TG overlap each other and the length of the shared cM from the candidate to the reference person is 15 cM or higher.

    While the TG is clearly based on the matches to my 2nd cousin, am I correct in thinking that all of the candidates are also genetic relatives to my father? My thinking is that the overlapping segments of shared DNA create a “bridge” from one end of the group to the other. Is this a correct interpretation? I have included the data showing the overlap below.

    Reference person for comparison (2nd cousin twice removed)
    Person 1 (3rd cousin once removed) – start: 53,578,630, end: 92,397,523 (22.9285 cM)
    Person 2 (unknown relationship) – start: 44,728,485, end: 74,729,886 (18.7368 cM)
    Person 3 (unknown relationship) – start: 40,502,153, end: 53,457,739 (16.8972 cM)
    Person 4 (unknown relationship) – start: 25,787,939, end: 42,780,201 (16.0449 cM)
    Person 5 (unknown relationship) – start: 14,814,620, end: 40,799,348 (29.4996 cM)
    Person 6 (unknown relationship) – start: 4,479,173, end: 21,874,977 (32.1783 cM)
    Person 7 (unknown relationship) – start: 5,436,722, end: 17,441,351 (23.9817 cM)
    Person 8 (unknown relationship) – start: 5,591,315, end: 17,441,351 (23.6205 cM)
    Person 9 (unknown relationship) – start: 2,580,494, end: 10,557,244 (18.7156 cM)
    Person 10 (unknown relationship) – start: 2,415,569, end: 10,488,442 (19.0706 cM)
    Person 11 (unknown relationship) – start: 1,921,500, end: 9,391,113 (18.5134 cM)
    Person 12 (my father) – start: 785,475, end: 7,883,772 (19.3229 cM)


    • Erik, This is stretching the TG too far. From 0.8 to 92Mbp is probably the DNA of several ancestors of the reference person. Whenever there are shared segments that don’t overlap by more than 7cM (a guestimated guideline based on my experience), there is the very real probability that they come from different ancestors. I would divide up what you have into at least 3 different TGs: .8 to 21.9Mbp; 24.8 to 74.7Mbp and 74.7 to 92.4Mbp – maybe more. Are these overlapping or Triangulated shared segments – there is a big difference (overlapping is not Triangulation). For instance do 2 and 3 match, do 3 and 4 match; I’m pretty sure 4 and 6 don’t match; nor do 5 and 9 (they could be from very different ancestors). Unless you have Triangulated all these people, some could be on different sides of the reference person – I’m guessing, some are. How many of these people does your father match? It’s not certain that he is even in a TG with them… Jim


  2. In my previous post, I had neglected to mention that the second cousin, third cousin, and my father all have a known common ancestor. That is an important element to include in the discussion.


    • Erik, A second cousin shares 2 of 8 great-grandparents. So the matches with persons 4 to 6 could easily be with different ancestry. Have you compared each of these Persons with each of the others? This is an essential step in creating a TG. Jim


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s