Shared Segment Analysis: When a Match is both Paternal & Maternal

I'm working in matches at FamilyTreeDNA this morning and was looking at my brother, Michael's autosomal matches. Each of us have about 5 or so matches that show as matching our paternal & our maternal side.  My Mom, my paternal aunt,  3 of my siblings, and myself have our tests at FTDNA. So I gathered the segment data for one of Michael's matches that I'll call P. Smith just to get an idea of what was happening.  I then checked the match lists of my Mom, my paternal aunt, my siblings, my own match list,  and also my daughter's match list.  I was interested in seeing the size of the segments, how many of us had both or none and which segments each of us had.   I also wanted to know if my Aunt also had the segment that my brother had received from Dad.  Because of visual phasing and overlapping/shared/ICW matches I have a pretty good idea of the family groups that these segments were from but to simplify things I've only identified which of my grandparents passed on these segments.  

My brother was the only one of us tested that got both segments.  My sister got the segment from dad that was passed to him by his mother.  My other brother, and I got the segment from Mom that was passed to her by her dad.   My paternal Aunt did in fact have the same segment that Brother 1 received.  I have passed the segment of my MGF on to my daughter as well.  

From looking at the matches tree and also at share/ICW matches it appears that the PGM segment goes back along the lines of my Morrow/Sutton on back to the Knox Craig lines.  Not 100% certain but that is currently my hypothesis.  That would mean this segment is back around the 5th Great Grandparent level for me.  (screen snip from my working tree shown below)

The segment which my Mom passed to us(from our MGF) appears to be from my Mom's PGM's lines and from a person/couple who was born in the mid 1700s.

As a big fan of DNAPainter for analysis of dna matches, my next step is adding the matching segments to the profiles for each of the tests.  This is a simple process for the tests that just have one segment but has a few extra steps when there are multiple lines involved.  What I could do would be to add both segments under one group and then move the other segment to the appropriate group but I prefer to do it a different way.  I add just the one segment and then click on the option to add additional segments.  For me it works better that way.

It's important to note when we share more than one segment with a match, we cannot assume that both segments are thru the same lines without further evidence.  This drives home the importance of testing as many siblings and aunt/uncles/grandparents as possible.  Also I'm very thankful that this match was on a site that has a chromosome browser and that I had a good mix of tests with which to compare.

Working AncestryDNA circles and a few observations

My sister and I have both tested at AncestryDNA.  When I am working on a segment and it is pointing in the direction of a specific ancestor, I always check my AncestryDNA circles to see if I have a Circle for that particular ancestor or ancestor couple.  I work the circles.  What do I mean by work?  I have a spreadsheet template I use.  I begin by adding the Ancestor's name at the top.  The spreadsheet has columns for:

• Ancestry User Name
• Match or Match to Circle
• Estimated cMs/segments
• Group(Child thru which they descend)
• GEDmatch #(known and added as they upload)

I add the info for each circle I work and this helps me to track not only who matches me but the descendant pathways.  Also and probably the most important thing for me is that I can see at a glance which tester would be the most valuable to have their info at GEDmatch.   I generally write those who Ancestry estimates match me on multiple segments or at amounts greater than 15 cMs.  I ask them to consider uploading to GEDmatch if they are not already there and if they have uploaded, how to identify them among my many matches.  I want to focus on these larger matches for a number of reasons. The circle membership gives me a pretty good idea of where they could match me. It is easy to get overwhelmed by all the matches but I feel if you focus on the larger ones, those will help in identifying others.   I also like to look at the smaller matches and see how their estimated segment matches are in comparison with my sister's results.  If they match her at amounts over 15cMs I also message them. 

Stay focused on the larger matches when trying to get folks to GEDmatch.  However do not forget the randomness of DNA.  Siblings can vary greatly in what matches they have.

My sister and I are a great example of that.

Since Mom tested I was able to make phased kits(thanks to GEDmatch tools) which show the matches that each of us got from Dad(who passed before we began DNA testing.)  I have compared my Paternal Phased Kit to my sister, Becca's Paternal Phased Kit and also compared each of our Maternal Phased Kits.   Below are the results

Comparing Paternal Phased for Marie and Paternal Phased for Becca(Full siblings)

(Chr 1-22, X not included)

Largest segment = 134.6 cM

Total of segments > 7 cM = 1,526.2 cM

33 matching segments

Estimated number of generations to MRCA = 1.6

Comparing Maternal Phased for Marie and Maternal Phased for Becca(Full siblings)

(Chr 1-22, X not included)

Largest segment = 145.6 cM

Total of segments > 7 cM = 2,513.8 cM

50 matching segments

Estimated number of generations to MRCA = 1.3

So you can see that on our Paternal lines we appear to have gotten a greater variety, while Mom seems to have given us more of the same.  

Table below shows another example of the randomness. The amounts shown are those estimated by Ancestry and they tend to be slightly higher at GEDmatch.

Comparing segment matches we have to
members of a Paternal Ancestor Circle.