Well here's another tutorial type post talking about matching DNA and the like. In an earlier post, I explained how I went about finding my Great-Grandparents. Where that leaves off, is where my Great-Grandparents had seven children.
If each of those seven children married and had three children, then there are 21 people who could potentially be my grandparent. If each of those 21 people married and had three children, then there are 63 people who could potentially be my parent!
Well - for starters - we have to build the family tree and find all of those people. I did this with a lot of help from my friend, Kristen. (Shout out to Kristen!) Building this tree required extensive searches on Google, Facebook, Newspapers.com, obituary searches, findagrave.com, Ancestry.com, beenverified.com, and anywhere else you can think of. There were days when I started my research before dawn, and was still researching well past midnight. Since Kristen is on the other side of the country from me, and we were messaging each other while researching, I know that she was often putting as many hours into research as I was - often she spent even more hours than me, as she was adding people to the family tree on the days I was at my day job.
I believe we spent a total of about six to seven months building the mirror tree for my closest match, which included most of my match's cousins, and then another month focusing on the descendants of my Great-Grandparents (after determining which of the ancestors in that mirror tree are my Great-Grandparents).
All that just to say, yeah, that was challenging.
With the tree built and my Great-Grandparents identified, the next step is to figure out which of the cousins in this huge tree is my father.
I considered buying birth announcements - you know the cards that announce, "It's A Girl!", and send them out to all of the cousins. Maybe someone would come forward. Oh but, maybe not. And maybe that would cause problems... which was the last thing I wanted to do.
Since we're looking for my father, we know that the first thing we can do is exclude all of the female cousins. We know when I was born, so we can also exclude all of the male cousins who are too young to be my father. I was told that my father was around the same age as my mother, so will focus on the male cousins around that age, but not exclude the older male cousins (not just yet).
With all that narrowing down, the huge list of cousins was brought down to about 12 males who could potentially be my father. Two of them deceased. A couple of them were married before I was conceived and have children older than me. (Those are the ones my "It's A Girl" card might have caused problems for.)
Of twelve men, eleven are cousins and one is my father - but what would set my father apart?
All twelve of those men are descendants of my Great-Grandparents. My father, however, is also the grandson of my *other* set of Great-Grandparents... the set of Great-Grandparents I hadn't identified yet.
The following diagram demonstrates the difference between the trees of my father and the cousin who is "Not the Father". In this tree, my ancestors are in blue, the ancestors I have already identified with DNA are outlined in red, and my genetic relatives are in yellow.
In the above diagram, I am not genetically related to the relatives in gray.
Notice the yellow "Not the Father". One of his parents is a descendant of my identified ancestors (this is what makes us cousins), but his other parent (in gray, labeled "Spouse of Aunt/Uncle") is not genetically related to me. Also, his cousins on the other side of his tree (in grey) are not genetically related to me.
Please note that the "Not the Father" person in this diagram is representative of all of the eleven cousins who are not my father. Showing all of them in one diagram would make it too large, plus you can place any of them into this spot in this diagram, and it would be the same. So it doesn't matter if I'm looking at a group of two cousins or twenty, the same principle applies.
The difference between "MY FATHER" and "NOT THE FATHER", is that in "My Father"'s family tree, I am genetically related to all of his ancestors and all of his cousins on both sides of his tree.
So, in trying to determine which of the 12 male cousins are my father, I need to build family trees for each of them. This tree then, must be much bigger than just the descendants of my Great-Grandparents. The tree needs to include the spouses of each of the descendants, as well as the ancestors of each of the spouses.
The following diagram demonstrates the cousins who had taken a DNA test to confirm who my grandparents were (2nd cousins circled in red) and how they are related.
In my case, my father was an only child, so I didn't need to go further than confirming my grandparents. Had my father had brothers, there would still be a question of which brother is my father. The brothers would have the same ancestors, so the ancestor tree isn't going to help in making this determination.
I would guess that the best way to determine which brother is your father is to ask them to take a DNA test. If they are unavailable or unwilling, you may ask their children to test. If you still can't get someone to agree to test, their grandchildren may be a possibility - but keep in mind that the range of possible matching DNA widens with each generation removed, so the results of grandchildren will not be as conclusive as the results of testing the men themselves.
The following diagram shows the average amount of expected DNA match for each relationship.
Numbers from: http://isogg.org/wiki/Autosomal_DNA_statistics
Here are the ranges reported on that site:
Father: 3400 cM (50%)
Aunt/Uncle: ave 1703 cM (25%), range 121-2227 cM
Half-Sibling: ave 1731 cM (25%), range 787-2134 cM
1st Cousin: ave 881 cM (12.5%), range 83-1559 cM
Half-Niece/Nephew: ave 892 cM (12.5%), range 540-1348 cM
1st Cousin Once Removed: ave 440 cM (6.25%), range 54-903 cM
Considering the reported ranges, the best way to be certain is to have the brothers take a DNA test. Since there is a chance that a half-sibling's results could be at the low end of the range, or that a 1st cousin's results could be at the high end of the range, such results could still leave questions. Again, if the brothers are unable or unwilling, having their kids test will be the next best option. You may find at this point that you have to rely on circumstantial evidence, such as asking which brother was at the right place at the right time.