How to Triangulate

Here is a 3-step process for Triangulation: Collect, Arrange, Compare/Group.

  1. Collect all the Match-segments you can. I recommend testing at all three companies (23andMe, FTDNA, and AncestryDNA), and using GEDmatch. But, wherever you test, get all of your segments into a spreadsheet. If you are using more than one company, you need to download, and then arrange, the data in the same format as your spreadsheet. Downloading/arranging is best when starting a new spreadsheet. Downloading avoids typing errors, but direct typing is sometimes easier for updates. I recommend deleting all segments under 7cM – most of them will be IBC/IBS (false segments) anyway, and even the ones which may be IBD are very difficult to confirm as such. You are much better off doing as much Triangulation as you can with segments over 7cM (or use a 10cM threshold if you wish), and then adding smaller segments back in later, if you want to analyze them. NB: Some of your closer Matches will share multiple segments with you – each segment must be entered as a separate row in your spreadsheet. The minimum requirement for a Triangulation with a spreadsheet includes columns for MatchName, Chromosome, SegmentStartLocation, SengmentEndLocation, cMs and TG. Most of us also have columns for SNPs, company, testee, TG, and any other information of interest to you. Perhaps I need a separate blog post about spreadsheets… ;>j
  1. Arrange the segments by sorting the entire spreadsheet (Cntr-A) by Chromosome and Segment StartLocation. This is one sort with two levels – the Chromosome column is the first level. This puts all of your segments in order – from the first one on Chromosome 1 to the last one on Chromosome 23 (for sorting purposes I recommend changing Chromosome X to 23 or 23X so it will sort after 22). This serves the purpose of putting overlapping segments close to each other in the spreadsheet where they are easy to compare.
  1. Compare/Group overlapping segments. All of these segments are shared segments with you. So with segments that overlap each other, you want to know if they match each other at this location. If so this is Triangulation. This comparison is done a little differently at each company, but the goal is the same: two segments either match each other, or they don’t (or there isn’t enough overlapping segment information to determine a match). All the Matches who match each other will form a Triangulated Group, on one chromosome – call this TG A (or any other name you want). Go through the same process with the segments who didn’t match TG A. They will often match each other and will form a second, overlapping TG, on the other chromosome – call this TG B. [Remember you have two of each numbered chromosome.] So to review, and put it all a different way: All of your segments (every row of your spreadsheet) will go into one of 4 categories:
  • TG A [the first one with segments which match each other]
  • TG B [the other, overlapping, one with segments which match each other]
  • IBC/IBS [the segments don’t match either TG A or TG B]
  • Undetermined [there are not enough segments to form both TG A and TG B                            and/or there isn’t enough overlapping data to determine a match.]
  • NB: None of the segments in TG A should match any of the segments in TG B.
  1. At GEDmatch – the comparisons are easy. Just compare two kit numbers using the one-to-one utility to see if they match each other on the appropriate segment. The ones that do are Triangulated. You may also use the Tier1 Triangulation utility or the Segment utility. I prefer using the one-to-one utility and Chrome.
  1. At 23andMe you have several different utilities:
  • Family Inheritance: Advanced lets you compare up to 5 Matches at a time. You may also request a spreadsheet of all your shared segments; sort that by chromosome and SegmentStart, and check to see if two of your Matches match each other. The ones that do are Triangulated.
  • Countries of Ancestry: Sort a Match’s spreadsheet by chromosome and SegmentStart, search for your own name, and highlight the overlapping segments. The Matches on this highlighted list who are also on overlapping segments in your spreadsheet are Triangulated (the CoA spreadsheet confirms the match between two of your Matches)
  1. At FTDNA it’s a little trickier, because they don’t have a utility to compare two of your Matches. So the most positive method is to contact the Matches and ask them to confirm if they match your overlapping Matches, or not. The ones that do are Triangulated. An almost-as-good alternative is to use the InCommonWith utility. Look for the 2-squigley-arrows icon next to a Match’s name, click that, and select In Common With to get a list of your Matches who also match the Match you started with. Compare that list of Matches with the list of list of Matches with overlapping segments in your spreadsheet. Matches on both lists are considered to be Triangulated. Although this is not a foolproof method, it works most of the time. And if you find three or four ICW Matches in the same TG, the odds are much closer to 100%. Remember, every segment in your spreadsheet must go in one TG or the other, or be IBC/IBS, or be undetermined. If a particular Match, in one TG, is critical to your analysis, then try hard to confirm the Triangulation by contacting the Matches.
  1. AncestryDNA has no DNA analysis utilities. You need to convince your Matches to upload their raw data to GEDmatch (for free) or FTDNA (for a fee), and see the paragraphs above.

Comments to improve this blog post are welcomed.


10 Segmentology: How to Triangulate; Jim Bartlett 20150511

46 thoughts on “How to Triangulate

  1. Regarding triangulation at. FTDNA is using the matrix tool not an option? …load the two that match you in the tool to see if they match each other. also having a parent’s kit to determine side of family on which you match is extremely helpful.


  2. The FTDNA matrix tool is a different way of looking at In Common With, and it can be used instead. A parent’s kit should not be used as one “side” of the Triangulation. After the Triangulated Group is establish, a parent’s kit is invaluable in assigning the TG to that side (parent).


  3. Thank you so much for putting this out for all of us to read. The fact that there are two of each chromosome, and that not everyone that looks like a match really is, is not well explained in most of the literature I have read.
    I look forward to all your future posts. Thank you again.


  4. if I match someone on multiple chromosomes, in order to triangulate – do I need to have a 3rd person that also matches on all the chromosomes, or just one??


    • David,

      Each shared segment is potentially from a different Common Ancestor, although usually multiple segments from one Match (usually a close relative), will be from the same, close, Common Ancestor. Each shared segment in a different location, on a different chromosome or not, should be part of a unique Triangulated Group. You need at least one other Match for each one. It will usually be different Matches for different TGs, but it could be the same one. The ones who also match on multiple TGs tend to be very close cousins/relatives. So you need just one, for each Triangulated Group.


  5. wow – thank you for the quick response. so to paraphrase, I need just one chromosome; but it’s worth creating a triangulated group for each to establish other possible CA’s. I very much appreciate your answer. it’s helped my understanding quite a bit.


  6. I have another question. can I triangulate using 2 knowns? in other words, can I use myself and my father and one other unknown person. then if that person matches me and that person matches my father, then I’m triangulated??


    • Jim – I was about to ask this question but remembered that I had asked it before. if you have a moment, can you respond to above (reposted below as well)??

      “I have another question. can I triangulate using 2 knowns? in other words, can I use myself and my father and one other unknown person. then if that person matches me and that person matches my father, then I’m triangulated??”


      • David

        You can do that (use yourself and a parent), but take care. Since you and a parent are fully identical, it like just sharing a segment with someone, and knowing it’s on one parent’s side. If the shared segment is less than 10cM, it might be IBS (and your parent may still match). So you are much better off triangulating with two widely separated cousins – such triangulation is almost always IBD with segments over 7 cM.

        Jim –



  7. I downloaded all of my FTDNA matches into Excel, sorted by cM and eliminated everything under 5.0, then sorted the way you recommended. I had previously sent all of my Ancestry and FTDNA to GEDMatch. Next I asked GEDMatch for a one-to-many and printed the full list of my 1,500 matches. I don’t need to concern myself with the lack of raw data from Ancestry since GEDMatch provides it. At that point things get tedious. Using the “two kits match” utility, I can get a list of other matches in a family group. I finish up that list with a one-to-one match and enter the data into the appropriate chromosome/start/end spot in my spreadsheet. While this sounds as if it makes sense to me, and the results look reasonable for known cousins, I would appreciate your review. I don’t want to spend hundreds of hours going down the wrong path, and you are the only person I trust with an answer.
    Don Sheaffer


    • Don

      What you are doing is correct. This is the collect and arrange steps. Now you need to group them. Determine which overlapping segments match each other. In general, for each area on a chromosome you’ll form two groups – on for each parent (although you may not know which parent, yet).


      • Jim, Thanks for taking time out of your busy schedule to reply. That explains why I got hits on the same chromosome at the same segment start location for both paternal and maternal cousins. At some point along that segment they must have crossed, right? Don

        Liked by 1 person

      • Don I’m not sure what you mean by crossed. Segments do not “cross”. Segments are formed between crossover points – when a parent recombines his or her two chromosomes to make a new chromosome for a child.

        Jim – Sent from my iPhone – FaceTime!



  8. Pingback: Shared Chromosomes and Triangulation Part 1

  9. Hi Jim… I’m using all segments from Gedmatch and adding additional matches from FTDNA then using a filtering out the smaller segments. I have a few matches from FTDNA which comprise of several small segments only. Should I consider these as true matches and when are they useful?


    • Jeanette, Yes, you can use them if you are sure the shared segment is IBD. I don’t know of any way to insure short segments are IBD. The algorithms can make short segments “look” like they match. So be careful, and ready to “delete” if you get contrary evidence…

      Jim – Sent from my iPhone – FaceTime!



  10. Thanx a bunch! That brings me to another question. These matches don’t qualify using the gedmatch criteria and probably would not make my list on Ancestry. Why does FTDNA consider them as matches? Is it simply because totaled they meet their criteria as a match? I guess no algorithm is perfect in genealogy. I think I’m just trying to understand the logic of them being on my FTDNA match list before I delete them.


  11. Checking on when I can declare victory. Using as my focus, I have identified an area where there are 4 people who match each other and myself. 2 of these people had posted genealogy trees that showed a match based on the tree. I contacted the other two. The first one to get back showed the same genealogy tree match.

    All the shared segments are about 12 cM. Each person has a total match with me in excess of 30 cM.

    The hypothesized relationship is a 4th cousin plus a 5th cousin (my paternal grandparents were 2nd cousins, once removed, from the same couple involved in this match).

    Time to declare a successful triangulation and move on to the next challenge?


  12. Dear Jim,

    I have one that is baffling my mind! In doing research to find my father, i came across a NPE. My mothers brother and sister are both half siblings. I have my mother’s other sister’s daughter test (my cousin) All of my Aunts and uncles have the same mother, this i am sure of, its the father in question. How do i triangulate the four of them to see if my cousin’s mother is a full sibling match to either my mom, or her brother and sister. I suspect my cousin’s mother (who couldnt test) is my mom’s full sibling. I just cant wrap my head around proving it. They all triangulate on my grandmothers side, How do I see the difference on the paternal side. Sorry if I am messing this up…
    All have gedmatch, and FTDNA.
    Feel free to email me directly


    • Melanie,

      You triangulate folks by finding common Matches on overlapping segments. Perhaps make a spreadsheet of all the Matches and segments for each one – include a column for the initials of the base person in each case. Then combine the four spreadsheets and sort on Chromosome plus Start Location and look down the list for Matches who appear near each other.
      A different plan is to upload all of their kits to GEDmatch and then compare them to each other. Full siblings will share twice the DNA as half-siblings.


      • Thanks Jim,
        Its a cousin match though. My mother and this cousin (who’s Mom I hope is my mom’s full sibling) share 1600 cms, and 160cms of it are an X. My mothers half siblings match this cousin (or their niece) at about 1100 and 1000 cms, and only have about 50 cms of the X. So I think that her mother would be a full sibling match to my mother. All are uploaded to Gedmatch. Mom – A690123, Laura (Cousin) A505275. Patty -A982500, and Eddie F412372.


  13. Hi Jim,

    Many thanks for this blog! Your clear, simple explanations are a great reference–I’ve read each post more than once and I look forward to future posts!

    I am experimenting with forming TGs on Chromosome 1 and I have a couple of questions. I created a spreadsheet of my GEDMatch matches, sorted by chromosome and segment start/end points. I then created groups (large for now) of the overlapping segments on each chromosome.

    1. I have results for both my parents. You mentioned forming triangulation groups first and then using parental results to assign the groups to maternal or paternal. I have results for both my parents and intuitively it seems that I should separate out maternal and paternal matches first and then create triangulation groups within each of those lists. Are the initial designations of TG A and TG B to separate out the maternal and paternal chromosomes? (I understand that I don’t use myself and a parent to triangulate, but does it make sense to make a first division of segments by separating my maternal and paternal matches?)

    2. I have a fairly large number of segments on Chromosome 1, 39/156, or 25%, that don’t match either parent. All of the non-matches are under 9.0 cM. Are those segments IBS and is that number typical?

    Warm regards,
    Terri Jensen


    • Terri – Thanks for the positive reinforcement. Triangulation is designed for everyone, and all you have to do is compare segments to see which ones match each other. This part is fairly straightforward, creating TG-As and TG-Bs. The next step is using genealogy to assign the TGs to Maternal and Paternal sides. When you know either parent, or both, you can go directly to TG-M and TG-P.

      1. As you form TGs, if the segments also match one parent or the other, you know directly if the TG should be TG-M or TG-P. And as you point out, all the segments that also match your mother can be set aside for TGs on the Maternal side (same for matches with your father on his side). In this case all that is then needed is to determine where one TG stops and the next one starts (remembering that segment ends are a little fuzzy, so there may be a small overlap; and/or there may be a gap between TGs). The key is that no segment in one TG will match a segment in another TG (except close cousins may share large segments which span several TGs). The order things are done is not important – each TG comes from a specific Common Ancestor. Whether you figure out which side the segments are on first and then group them into TGs, or form TGs first and then figure out which side – it doesn’t matter.

      2. You are also correct that a high percentage of segments at GEDmatch from 7 to 9cM are false. GEDmatch has no screening method in their algorithm (like other companies have), so all of these segments are presented to you as a “match”. As a tip, I usually drop the matching criteria to 500 SNPs and 5cM to check these segments against a parent – sometimes they have 690 SNPs (just under the 700 threshold), or 6.9cM (just under the 7.0 threshold). Those that then match a parent with this lower threshold, should be compared, using the lower threshold, against other segments in the appropriate TG – most will match, and may then be included in the TG.

      3. And one other observation. Sometimes I write about comparing Matches. I should always say compare segments. I have found a small number of Matches with two segments, and one segment triangulates on one side, while the other segment triangulates on the other side. In other words those Matches will share two different Common Ancestors with you – one on each side. This can also happen on the same side, which highlights the concept that the Common Ancestor for each TG has to be determined independently. If a Match shares multiple segments with you they will usually be from the same Common Ancestor, but not always. So be aware.


      • Thank you for your reply, Jim. This all makes sense.

        With respect to #2 above, yes, that worked! In my first TG on Chr 1, lowering the threshold as you suggested showed the “no matches” to be maternal matches (which I would have expected because those “no matches” matched others in the TG), just under the original threshold. That also cleared up why some of the maternal matches in that TG on Chr 1 didn’t match each other (they all matched when I lowered the threshold). I confirmed the matches using my maternal phased results from GEDMatch (with lower threshold), too.

        I think I dived in too quickly to an area of short segments where it’s trickier to understand what’s going on. Two of the people in the above maternal TG have also tested on FTDNA, where they match me (but not my Mom), and aren’t ICW each other. But one is ICW a “new” FTDNA person (not on GEDMatch) who matches both Mom and me within the TG, on a 22.56 cM segment, so that’s a possible lead. Though most of the segments in this TG are small and the CA probably distant, this particular group is exciting to me because I know one member of the group to be from somewhere on my Mom’s “recent immigrant” side, where I have few matches and am at the brick wall.

        This is fun! With experience I’m sure more will become clear. Time to work through some more chromosomes and TGs with longer segments and a known CA or two. Thanks again!


  14. Your site is a breath of fresh air, a bringer of clarity to the fog of genetics that swirls around us newbies.

    I need confirmation that if I share a specific segment over 10cM with two different matches on two different test sites that the three of us must triangulate on a common ancestoral line but not necessarily the same MRCA. True/false?

    As an example, a recently discovered 2nd cousin matches 5 of us close family in the followng way

    Start End
    My 1st cousin 1 24000000 49.8 cM 9144 SNPs
    Same cousin 27000000 33000000 7.6 cM 1294
    Another 1st cousin 1 8000000 19.4 cM 3630
    My brother 1000000 24000000 46.5 cM 8549
    Same brother 27000000 32000000 6.7 cM 993
    My son 1 15000000 33.0 cM 6270
    Me 1 24000000 49.8 cM 9139
    Me again 27000000 31000000 6.5 cM 892

    Now if I find matches on any other site including GEDmatch and DNAGedcom, that start on chromsome eight at 1 that they are definite triangulated matches if they exceed 10cM. If that chrm match is not the strongest match then it is still a decent confirmation.

    If the matches exceed 10 chr but don’t start and/or end at the same point on the chr are they still triangulation matches?

    If a different invidual matches me at two smaller fragments me within the start and end point of a larger segment, such as the fragment of chr 8 starting at 1 above, can I tentatively include that as valid match that has only been broken by recombination along their ancestral line?

    Thanks for your help. I’m expecting your response to enable me to sleep better. Not that I want to put any pressure on you or anything like that.

    Ian “McCallum”


    • Ian – a long complex example with a lot of questions. Some of your shared segments don’t have Chr # and both start and end locations. But if I get the gist of it, you have overlapping segments which match each other. And they are on the same side of your family. Then they should share a Common Ancestor with you. Some may be closer cousins (with closer MRCA), but all of the CAs will be on one ancestral line of yours.
      The shared segments do NOT need to start or end at the same location; and one can be included in another. They should overlap each other roughly at least 7cM by an eyeball estimation. The key is that they match each other. Hope this helps.
      Sweet dreams…


  15. Thanks for the reply Jim. It clarifies a number of questions that I had.

    It looks like I wasn’t clear enough about two important specifics (chr number #8 and start/end points), the latter compounded by the table’s format collapse.

    First specific, the examples are all how my recently discovered 2nd cousin (Karen) matches with these five male members of our family on chromosome 8, There are 5 of us: 2 of my 1st cousins, my brother and my son. That explains why there is so much consistency between the lines, for example four of the lines start right at, the beginning of chr8, @1.

    The analysis comes from 23andme.

    My larger question concerns a “third” party, one not included here. If that party has the same or a large portion of the same chr 8 fragment, can I then assume that this new match is also related to my 2nd cousin Karen, since she too matches me on that fragment? I assume so, but I need confirmation of that point so that I can confidently use it to produce a short list of matches to contact.

    I hope that clarification helps.

    Ian “McCallum”


    • Ian,
      Just having the same segment area is insufficient. Both your maternal and paternal Chr 8 have that area. The “third” party has to match at least two of the others in the group.


  16. Thanks Jim. So I assume then that anyone else who matches Karen and me on that fragment at the same lab can be used to triangulate with third parties, say on GEDmatch . Using close relatives is not absolutely necessary. Useful. Thanks. Keep up the excellent work.


  17. For my simple brain:
    a “match” is someone who shares a large enough amount of DNA with you to be identified as a potential relative.

    a “match segment” or “matching segments” are the pieces of that shared DNA, identified by a beginning and ending location number on a chromosome.

    “triangulation” is when three or more people who match each other (are matches) share the same matching segment.


  18. I have a group in my match list that I don’t know what to do with. I ran the triangulation function at Gedmatch and came up with the following on chr 1.

    A762411 A443210 3,310,659 11,502,057 16.6
    T093771 A762411 5,197,705 14,367,590 17.7
    T093771 A443210 5,465,256 11,502,057 10.7
    M485905 A443210 7,073,164 11,502,057 7.4
    A762411 M485905 7,073,164 14,362,549 13.9
    T093771 M485905 7,073,164 18,555,636 23.9
    A443210 A024222 7,408,904 11,717,263 7.4
    A762411 A024222 7,408,904 14,367,590 13.3
    T093771 A024222 7,408,904 18,299,732 22.3
    M485905 A024222 7,408,904 18,299,732 22.3
    A443210 M173836 9,457,890 11,719,611 4.1
    A443210 M173836 12,058,113 13,654,784 3.2
    A762411 M173836 10,407,606 14,365,065 8.9
    A024222 M173836 10,407,606 18,339,740 18
    M485905 M173836 10,407,606 18,555,636 18.9
    T093771 M173836 10,407,606 18,606,731 19

    There are 6 different people here that all match each other, and me of course.. But if I just take the common segment with all of them combined I only get a segment from 10,407.606 to 11.502,057 which is very small. How should I separate these into different groups? Or is each line a separate group?

    I have not yet contacted any of these people to find common ancestors. I am just trying to understand the concepts here.


    • Dean – We need to understand each other. When you say they “all match each other”, I don’t get it, because I don’t think the first one and the last one will match (it doesn’t look like they overlap by 7cM). So do you mean you compared each one to each of the others? Matching doesn’t mean they are all on the same chromosome – it means they are compared to each other and they share a segment with each other on Chr 1 in this area. So the concept here is to compare the first two to each other – if they match on the same segment area they are a TG; if they don’t they are in separate TGs. Then compare #3 with one or the other until it fits with one. As you go down the list, each succeed one will match one TG or the other TG or neither (in which case it’s a false segment – usually under 10cM. In the end, with the group above, you’ll probably get two TGs, but you might get one TG on one side and two adjacent ones on the other side. Hope this helps.


      • This list came from the triangulation group function at Gedmatch, so each line triangulates with me. This means that at the base level each line is a triangulation group on it’s own. I think I am understanding this part correctly; if not please correct me.

        When I say they all match each other, I mean that each kit matches every other kit at some point in this list. A762411 matches A443210 and T093771 and M485905 and so on for every other kit in the list. And carrying this on, every kit matches every other kit in the list. Every combination is there. Wouldn’t this mean they all match on the same chromosome of my pair of chromosome 1s?

        I guess it will probably become apparent what is going on here if I can get anyone to reply to email, but that’s proven pretty difficult so far.


  19. Dean, generally the TG is not just the small overlap of all shared segments; but is from the smallest start location to the largest end location. Generally… You had to get this segment, intact, from an ancestor in order to match each of the others in the TG.
    In looking over your list again, I am very surprised that GEDmatch reported a 3.2cM segment.


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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s