Cousins in Common

The prior article concerned DNA cousins in common.  Finding DNA cousins who match you and some of your other DNA cousins is not uncommon.  You must be careful in drawing inferences in these situations.

Finding cousins in common is a tool at FamilyTreeDNA and GedMatch.  You can manually compare cousins at 23andMe to ascertain if someone matches your cousin- as long as everyone is sharing genomes.  The glitch arises when two of your cousins match each other, but not on the same segments of DNA where you match each of them.

The query:  If three people share different segments of DNA with one another, are they related through a common ancestral line?

Answer:  Maybe, maybe not.

Unless the match is very close, I don't pursue cousins in common.  Early on in my genetic genealogy pursuits, I was overwhelmed with my mother's DNA matches.  They matched her and her other DNA cousins, though not on the same segments.  To this day, I have no direction in this undocumented branch of her family tree.  These cousins either live in or have recently immigrated from Eastern Europe.

Two years ago, my maternal third cousin appeared among the matches at 23andMe.  We share a pair of great great grandparents, John Preston (1857-1928) and Bridget Sheehey (1857-1916).  John was born in Dutchess County, New York to Irish parents.  Bridget was born in Ireland.  (I have no idea where in Ireland they lived.)

Here is the graph of his shared DNA with my mother- three segments:

With this new close cousin's DNA, I compared him to my mother's distant DNA cousins to round up some people that we could place in the Preston/Sheehey branch.


Comparison of my mother's distant DNA cousins from Eastern Europe revealed that this Irish Preston/Sheehey cousin matched a lot of them on multiple segments.  Lots of cousins in common, but from different branches of their respective trees.  To date, no DNA cousin from Eastern Europe matches my mother and this Irish Preston/Sheehey cousin on the same segments.

The point is to be cautious when looking at cousins in common who do not share the same DNA segments.

Cousins of Cousins

A DNA cousin at 23andMe shares identical DNA with my father and three of his close cousins- each from a different branch of the tree.  This is why you need to see the segment(s) shared between DNA cousins and you need to test your own close cousins to narrow down which branch of your tree may hold the common ancestor.

The match to my father will be in his Duryea and Cummings branch because the segment of DNA is shared with the cousin from descended from Abraham Brewer Duryea (1878-1944) and Nellie Cummings (1879-1965).

The segment shared between the DNA cousin and the cousin from the Uhl and Patschke branch could match anywhere in the cousin's entire family tree.

Same for my cousin from the Cook and Neal branch- the common ancestors could be anywhere in his tree (but not too far- they may be about fourth cousins).  Matching my father does not narrow down the possibilities because my father's shared segment with this DNA cousin can be attributed to another branch.

I cringe when people write to me about DNA cousins in common when they don't match on the same segment.

DNA shared with a Third Cousin

In a prior post, we were introduced to a third cousin who reached out to me.  The common ancestors are a pair of German great great grandparents who settled in Newark, New Jersey: David Uhl (1834-1884) and Clara Patschke (1840-1914).  I featured pictures of the heirlooms from these ancestors- some passed down through my branch and others through his branch.

Just as the heirlooms from these ancestors were dispersed to the succeeding generations, so was DNA.  This cousin kindly added his DNA to the ongoing genetic genealogy studies of the family at 23andMe.

My father and his three siblings all share autosomal DNA with this third cousin.

 The amount of shared DNA for three of the siblings is on the high end, while the other sibling is on the low end, sharing only 43 cM.

This picture represents where on each chromosome these 3rd cousins match.  The colored lines represent segments of DNA that traveled four generations from David Uhl and Clara Patschke to their great great grandchildren.

Many people- "DNA Cousins"- match my father and his siblings on these segments.  If these DNA Cousins also match the known third cousin, then we know to look into the branch of our family that holds David Uhl and Clara Patschke.  I will have information that the DNA Cousins do not, and vice versa, and we will break down brick walls.

When compared to my father and his two children (my sister and me), we see interesting results.  I inherited most of my father's DNA that he shares with this third cousin.  My sister received almost nothing:  13 cm over two segments.  This is such a small amount that a DNA testing company may not even recognize it as a match on its own.  Imagine if my sister were the only one who tested her DNA.  A match this small would likely be disregarded, yet it is fully solvable based on surnames and locations.

Two years ago, we compared DNA to another third cousin from a different branch- great great grandparents Calvin Cook (1826-1889) and Mary Neal (1830-1898).  The amount of shared DNA was lower than what we see with the Uhl/Patschke cousin, with a low of 70 cM to a high of 121 cM.  The next generation- my sister and me- inherited none of this DNA.

Chromosome Mapping

My genealogical research has been enhanced through the use of DNA studies.  Several relatives have kindly submitted their DNA to help me gather both known and unknown branches of the family tree.  Increasing numbers of people are testing their own DNA for different reasons.  Some methods of analyzing the results are provided by the testing companies themselves, while other tools are being developed by (extra)ordinary people.

This weekend I had the pleasure of learning about a new tool developed by Kitty Cooper:  Chromosome Mapper.  (This is for autosomal DNA results and not Y-DNA or mitochondrial DNA tests.)  I was introduced to the idea of colorfully mapping the chromosome by Jim Owston.  The purpose of mapping a chromosome is to visually identify which pieces of DNA came from which ancestor, thereby helping you identify ancestors in common with your genetic distant cousins.

[Here's the short technical aspect:  to use the Chromosome Mapper, you first need to have your autosomal DNA tested, obtain the results, and then identify ancestors in common with some of your matches.  From there, you need to create an Excel spreadsheet of the matches and the MRCA (most recent common ancestor).  If no close relatives have tested and you have not identified the MRCA of any match, then you cannot map your chromosomes.]

My father, David Lutter, has three cousins tested from his mother's side and one cousin from his father's side, so the common ancestors were already identified.  Plugging their numbers into the Chromosome Mapper produces a useful visual tool.

The colors represent areas of David's DNA that can be attributed to specific ancestors.
Chromosome Mapper at KittyMunson.com

When a distant genetic cousin matches David and one of his known close cousins on the colored areas shown above, we know which branch of the family holds the common ancestral line.

Ancestral fan chart (Family Tree Maker 2012)
DNA matches on the blue areas of David's DNA come from the Dunn/Winterton branch of his family tree.

Distant DNA cousins who match David on the orange areas of his chromosome map come from the Neil/Cook branch of his family tree.

DNA matches that fall on the red areas of David's chromosomes come from the Cummings/Duryea branch.

In the fan charts above, the females are indicated in pink and the males in blue.  The chart extends to David's great great great grandparents.  Please note the white boxes, indicating an unknown ancestral line.  It is entirely possible that the most recent common ancestor of David and a distant genetic match lies beyond this unknown area.

The Chromosome Mapper allows for a lot more ancestors and colors, so I plugged in my own matches using my mother's and father's matches.  (I omitted my paternal third cousin, once removed.  I share no DNA with him from our known common ancestors, Mary Neil and Calvin Cook.  Instead, we share a small segment that I inherited from my mother.  At this time, we do not know the common ancestors of my father's third cousin and my mother, but by testing my mother's cousins, we know that the match will be in her Joyce/ODonnell branch.)

Several chromosomes show two colors in the same area because they are showing both the maternal and the paternal sides of the chromosome.  I think that this is so neat to be able to visualize which areas of my DNA came from which ancestors- people I can never meet because they are long gone.

Both of my parents have tested their DNA, so most of my matches can be quickly assigned to either the paternal side or the maternal side.  A few people match me but neither parent, while several people match both my parents!  My grandparents are all deceased, so dividing my parent's matches into their paternal and maternal sides becomes challenging.  Testing cousins from different branches of the family tree narrows down the possible areas in which to look for a common ancestor to a distant genetic match.

While studying my Chromosome Map versus my father's Map, I realized that I will have overlapping or common areas of match on my maternal line.  This is because all of my mother's relatives who tested descend from the same set of ancestors, Delia Joyce and Patrick ODonnell.  23andMe provides a mapping function, limited to three comparisons.  Here is my map of my maternal matches (one person in each category).

23andMe
Areas of Jody's DNA in common with descendants of specified ancestors.

And here is the same mapping from Kitty's Chromosome Mapper:

Chromosome Mapper/KittyMunson.com
I specified "paternal" in my spreadsheet to achieve similar coloring to the 23andMe map.
Also, Chromosome Mapper allows input from several cousins to form a more complete map.
The dark blue on this map represents DNA from one cousin (compare to two below).

With the goal of narrowing down which branch of the family holds the Most Recent Common Ancestor, I would prefer to see the areas where I match the most distant generation (a set of great great grandparents/dark blue) dominate.  The longest segment appears instead of the shorter segment, regardless of generation.  Any DNA from these great grandparents (green) was passed down to me through their daughter (light blue), so those two colors actually represent the same branch.  The useful function is to further differentiate the DNA from these great grandparents (Delia Joyce and Patrick ODonnell) into my great grandfather (Frank ODonnell).  I can manipulate which information I put into the Chromosome Mapper.  By adding the areas of match for two cousins of my grandmother's generation, the DNA that they both share with me combines into extended dark blue areas, representing some of the DNA I inherited from my maternal great grandfather, Frank ODonnell.

Chromosome Mapper by Kitty Munson
Jody compared to two descendants of equal relation to this ancestral couple (cousins of Jody's grandmother).
This DNA tool visualizes all of their DNA into one cohesive color and segment.

Great work, Kitty.

Mystery Third Cousin of a Third Cousin

Inevitable at the DNA testing companies are those close matches who won't answer your requests/pleads for contact and information.  Here is one such match for my Known Third Cousin.

Nestled among my Third Cousin's highest matches (my father and his siblings) is this mystery cousin.  You can see that she lines up beautifully as a third cousin match.  I'm not saying that she matches my father or his siblings.  The match is to my Known Third Cousin- one of his sixteen great great grandparents is one of hers, or a sibling to hers.  Finding someone as close as a third cousin in the genetic database is a great boon to your research and helps you assign distant genetic cousins to a branch of your family tree.

23andMe Relative Finder/DNA Relatives
Top genetic matches for my Known Third Cousin

DNA Matching with a Third Cousin Revisited

DNA results are arriving for multiple family members who recently tested at 23andMe.  Processing and analyzing them is time-consuming.  Challenge Accepted.

First up for discussion is one of the first people to submit his DNA, my father's third cousin.  (Happy Birthday!)  Third cousins share a pair of great great grandparents; in this situation, Calvin Cook and Mary Neal, both born around 1830 in Morris County, New Jersey.  Once we reach the third cousin level of relation, the chances of sharing any identical DNA begin to decrease.  If third cousins do match, the percentage will be around 0.78% (watch the decimal) over two to six segments.

International Society of Genetic Genealogy

We lucked out and my father and his third cousin match within the predicted range:  0.93% over two segments.

23andMe autosomal DNA comparison between third cousins.
They share identical DNA on chromosomes 13 and 21.

The value in this information is uncovering which segments in the genome of my father and our third cousin could be attributed to Calvin Cook or Mary Neal.  As more people submit their DNA and compare their genome to ours, we can find matches.

23andMe autosomal DNA comparison
of two known third cousins
against two others who match both of them on the same identical area

These distant cousins match both my father and his third cousin on the same segment of their DNA.  This predicts that if we travel backwards in all of our family trees, we will find an identical branch, or the Most Recent Common Ancestor.

Finding the elusive Most Recent Common Ancestor is not easy.  Most distant cousins in the database do not have extensive family trees.

My father's three siblings tested their DNA.  Fortunately, all three also match their third cousin- with more DNA.  The amount of matching varies, but is still within a third cousin range.

23andMe autosomal DNA comparison
Relative Finder/DNA Relatives function
Four sets of known third cousins

23andMe autosomal comparison
Three siblings versus known third cousin.
The segments circled in red are segments newly identified as Cook/Neal genetic material.
All three siblings provide more matching DNA than the original comparison with my father.

My father's siblings share the two segments shared between my father and their third cousin, in addition to several other segments (marked in red in the screenshot above).  I can now identify these additional segments as Cook/Neal DNA in my father's siblings and their third cousin.  This enables me to identify more people in the database who share ancestry within our specific Cook/Neal branches.  This is especially important because Mary Neal is a "brick wall" in our family tree:  We do not have her parentage.  I suspect she is of Irish ancestry and am not surprised to find that most matches in common among my father, his siblings, and their third cousin are in Ireland.

Next, we consider the DNA shared between this third cousin and the next generation:  My sister and me.  Our relationship is called Third Cousins, Once Removed.

23andMe autosomal DNA comparison
between two siblings and known third cousin, once removed

I share no DNA with my third cousin, once removed, while my sister shares one small segment.  This is entirely possible and demonstrates how little DNA, if at all, you share with distant relatives.

Now here is where you can and will run into trouble with your DNA comparisons.

Let's say that my parents and their siblings were not available to test.  This is a very real situation for most of you reading this.  After finding no shared DNA between me and this third cousin, once removed, we might question our records and wonder if we had a non-paternal event in our lines.  Next my sister tests and are relieved and thrilled to find some shared DNA.  Looking at my sister's small segment shared with this third cousin, once removed, we would incorrectly conclude that this little segment is Cook and/or Neal DNA.  The segment shared between my sister and our third cousin is not from our father, but rather from our mother and is likely not Cook or Neal DNA.  (Leave open the possibility that our mother is related to Mary Neal, our father's great great grandmother.)

23andMe autosomal DNA comparison
Known paternal third cousin versus father/mother/child trio.
The green segments represent third cousin matching.
The blue segments represent a match to a distant cousin, passed on to the next generation.

Fascinating, isn't it?  Our father's cousin is also our mother's cousin!  Our father shares two large segments of DNA with his third cousin and we inherited none of it.  The small segment shared between my mother and my father's third cousin is inherited by my sister only.

Testing more cousins on our mother's side should help assign this double cousin to a maternal branch as well.  (Hint hint.)