Introducing the Borland Genetics Segment Lab

If I match Cousin A and Cousin B on chromosome 6, does that mean that we share a common ancestor?

The answer is a bit complicated!

First, the general answer:

General rule:  If you and cousin A and cousin B all match across a significantly overlapping span of the same copy of the same chromosome (called a segment), it is highly probable that you do so because the three of you all inherited that segment from a common ancestor.

But what does that mean?

First qualification:  It's not enough that the matches both match you on the same chromosome.  The matches must also match you over the same span of that chromosome.  Position on a chromosome is generally measured in MBP or Mega-Base-Pairs.  If Cousin A matches you on chromosome 6, but over the span of 23 MBP to 51 MBP, but Cousin B matches you over the span of 63 MBP to 81 MBP, this is not evidence that you and both matches share a common ancestor because these are segments to do not overlap.  How can I find out if the matching segments overlap?  The tool is called a "chromosome browser," and is offered by most testing companies (except Ancestry) and by third party austosomal DNA utilities such as those provided by Borland Genetics.

Second qualification:  The matches must both match you on the same copy of the chromosome.  It is vital to your understanding of genetic genealogy that you know that you have two copies of each chromosome, one paternal copy (inherited from your father) and one maternal copy (inherited from your mother).  This means that 50% of your DNA is inherited from your father and 50% from your mother.  But what if you match Cousin A on the paternal copy and Cousin B on the maternal copy of chromosome 6?  Then surely, the common location on the chromosome is coincidental and you did not inherit these segments from a common ancestor (excluding the rare possibility that your parents are related and by chance passed to you the exact same segment at this location).  Testing companies do not tell you which copy of your chromosomes a match lies on, so you have to use one of two techniques to determine whether your matches are on the same copy (be it your paternal or maternal copy).  The first and most common technique is called segment triangulation.  The process is simple, but it needs to be applied to each pair of matches you wish to investigate.  The inquiry is simply whether or not Cousin A and Cousin B also match one another along that same span of chromosome.  If they do, the matches are said to triangulate, and you all likely inherited the segment from a single common ancestor.  Otherwise, one match is paternal and the other maternal.  Another technique is called phasing.  If you have tested a parent or a child (or in some cases if enough more distant relatives), you can phase your DNA kit, which simply means splitting it into two separate DNA kits, one containing the paternal copy of your chromosomes and one containing the maternal copy of your chromosomes.  While this takes a bit more effort upfront, once you phase your data, you will never have to triangulate again.  Rather, the test is just whether the matches to both Cousin A and to Cousin B are to the same phased kit.

Third qualification:  Even if the matches are on the same span of the same copy of the same chromosome, this only creates a presumption that the shared segment is due to a common ancestor.  The larger the size of the shared segment (the portion of they chromosome where both Cousin A and Cousin B match you), the higher the likelihood that the shared segment is by common inheritance rather than coincidence.  When making comparisons using unphased kits, a "segment" of less than 7 cM is usually not the result of common inheritance, but rather the segment is what we call IBS (identical by state).  We can only make an inference regarding common ancestry with respect to segments that are IBD (identical by descent).  If you are using phased kits for your comparison, the same can be said for matches less than 5 cM.  It is strongly advised that you do NOT lower threshold settings below 7 cM for unphased comparisons or  below 5 cM for phased comparisons on sites like GEDmatch.

Fourth qualification:  Even if the shared segment is in fact due to inheritance, the common ancestor from which you all inherited the segment may be very distant.  Single-segment matches of less than 25 cM can often be impossible to place in your tree and could date back as far back as 1,000 years or more.  Sometimes this will be apparent, when you have a "pile-up" of matches on the segment that all triangulate.  A pile-up can consists of hundreds of matches in some circumstances.  You can see this visually using tools like the HIR Mapper at Borland Genetics.  When you have hundreds or thousands of matches on a single segment, this is a good indicator that the common ancestor lived long, long ago, prior to the genealogical era, and you will be unlikely to find a common ancestor between all your matches on the segment.  What is happening in this situation is called the "founder effect," where the distant founder of an endogamous population has propogated his or her DNA segment to a sizable percentage of the living descendants of the population via inter-marriage within that population (whereas in non-endogamous populations segments tend to "die out" since in each generation, half of a parent's genetic data is not transmitted to a child, and so there is a 50% chance that the child will not inherit the segment from the distant ancestor).  In an endogamous population, a segment may have become ubiquitous among the population, and you having inherited that segment is really just an indicator that you are descended from that population, although the segment did likely occur in an individual common ancestor that lived long before historical records.

To sum it up:

If you match cousin A and cousin B, and they also match each other across at least a 25 cM span of the same copy of chromosome 6, then yes, you inherited this segment from a common ancestor, and assuming you don’t match hundreds of other people on this same segment, the common ancestor likely lived within the genealogical time-frame, and with a bit of work, you can probably find that common ancestor, depending on the availability of paper genealogy resources in the region that ancestor lived.

Related video: