The Gray mystery was discovered when Renee's paternal aunt came up as a ZERO match, and her niece came up as a half strength match to her but a correct match to the paternal aunt. It appeared that of the 4 siblings that grew up thinking that Ken Gray was their father, in fact he was only the father of Ricki. Renee and her 2 brothers had a different father. Apparently Mother had an at least 6 year affair with the same person that produced 3 children. All this happened in the 1950's so it seems almost certain that while Mother must have suspected the possibility, she had no way knowing who the father was. The nominal family tree with the puzzle is shown below:
Step 1 is always to group all the DNA matches into exclusive groups using the shared match feature. I am looking for 4 independent groups that corresponding to the 4 grandparents families. In this case, the 4 groups were led by the Odom, Royalty, McCain and Price families. If there is test data from multiple vendors, it is important to pool the shared matches using surnames in common, so that there is complete grouping of the DNA matches.
Step 2 is to construct a family tree for each grouping, hopefully linking them back to the starting nominal family tree. Based on the detailed trees discussed later, the Odom and Royalty lines merged with J Coan Odom and Ida Mae Royalty, who had the right age to be grandparents. In order to inherit DNA from both lines, one of their sons had to be Renee's father. This was confirmed when the census record showed that the Odom sons and Mother were all living in Rotan TX, a town or 2-3000 in the Texas panhandle. Of the 3 brothers, one BB Odom showed a striking resemblance to Renee's brother Ross and he was a welder who travelled around the panhandle for work.
The 1950 census shows that the Odoms and Prices were neighbors at the start of the 50' when all 3 kids were born.
The third independent group, McCain therefore had to be the maternal grandfather, which is supported by the low DNA matches to maternal cousins. The McCains lived in the same county, about 30 miles from Rotan.
The resulting family tree is shown below:
The detailed trees for each DNA group is constructed by starting with any public trees associated with the individual DNA matches looking for surnames in common within the group. The next step is to research the trees for each surname to find out how they are linked and where different surnames merge through children. A powerful trick is to use a search in the public trees for key names. Often it is a question of just systematically going through each person in the tree building out the tree. The next step is to add the matches that do not have public trees, by searching for the individual matches and hopefully finding parent names to build the tree up from the bottom. The challenges are that there is no QA on the trees, and lots of people just copy existing trees that can just propagate errors. Many individuals swap first and middle names, change spelling, and anglicize names. Many families use a limited number of first names resulting in errors and confusion. Finally, all it takes is a Non Parental Event (NPE) to propagate DNA without any family tree record.
Each of the trees linking the DNA matches are shown below, along with some of the key relationship DNA match ranges. By comparing the DNA match to the relationship range, the proposed can be evaluated.
Maternal grandfather "Charles Edgar McCain" has 11 DNA matches with relationships that fit , and he lives in the same county.
Maternal grandmother "Lois Price" has 4 DNA matches with relationships that fit so is the only member of the original nominal tree.
Paternal grandfather must be a "J Coan Odom" with 6 DNA matches with relationships that fit, living in Rotan TX.
Paternal grandmother 1 must be a "Ida Mae Royaty" with 7 DNA matches with relationships that fit, living in Rotan TX.
Father is one of the sons of J Coan and Ida Mae, most likely "BB Odom".
The complete "Possible" tree is shown below between the maternal and paternal trees.