Genetic Genealogy: Chapter 7

 Chapter 7: Incorporating DNA Testing in a Family Study

Questions 1-7 refer to the family tree of Henry Smith in Appendix A. Recipients of mtDNA of #2 unknown are circled; recipients of #1 Henry Smith's Y-DNA are highlighted.

Recipients of mtDNA of #2 unknown are circled; of Y-DNA of #1 Henry Smith are highlighted

1. Assuming only end-of-line descendants are still living, is or are there any living who inherited the mtDNA of the unknown spouse #2 of Henry Smith #1?

Yes, #41 Tommy Curtis.

2. Assume a living person is found and agrees to test and that the results indicate the mtDNA haplogroup is U5b1c1. Does this confirm or refute the family legend that Henry's unknown spouse #2 was Choctaw?

Neither. U5b1c1 is not a known Native American mt haplotype, so that rules out unknown's mother or her umbilical line being NA. Since that is the only line tested, the rest of her inheritance is as yet unknown.

3. Assuming only end-of-line descendants are living, is there anyone living who may have inherited the Y-DNA of Henry Smith #1?

Yes, Ira Smith $43, Louis Smith #44, Robert Smith #33, Bobby Jack Smith #45, and Alfred Smith Jr. #46. 

4. George Smith #4 is believed to be the son of Henry Smith #1. Ira Smith #43 and Robert Smith #33 take a 37-marker Y-DNA STR test. Robert also took a SNP test. Their haplogroups do not match; are they related? Use https://isogg.org/tree/2015/ISOGG_HapgrpR15.html rather than the link in the book.

Since Robert took a SNP test, his haplogroup is more precisely named, but they are both from the same branch. However, this alone is not enough to support the hypothesis that George #4 is the son of Henry #1.

STR-markers table p. 118

Correlate the STR-marker values from the table with the "Descendants of Henry Smith" tree.

It is evident that if the tree is correct, some differences have arisen over time in the Y-DNA STR markers.

5. Ira and Robert do not match on markers DYS464 and CDY. Given that differing on up to 3 markers in a 37-marker test indicates relatedness, does this result prove that George Smith #4 is the son of Henry Smith #1?

No. They are related on the paternal line, but the low number of markers tested is not conclusive. Testing more known Y line descendants of Henry would help, as would both more STR testing and SNP testing such as the Big Y, which is the gold standard.

6. 

Marker significance for Y-DNA of Henry Smith #1?

Describe the significance of the data above to determine if George Smith #4 is the son of Henry Smith #1. All other men in the Smith Surname Project with similar haplotypes have marker DYS607=15.

• a) All tested descendants of Henry #1 share DYS607=14. What does this tell us about Henry #1?

Henry #1 probably was positive for DYS607=14.

• b) All tested descendants of Henry #1 share DYS607=14. What does this tell us about Ira #43?

That he may be a descendant of Henry #1 since he also shares this marker.

• c) All tested descendants of Henry Jr. #5 have marker DYS464=12-15-16-16. What does this tell us about a potential link between Henry #1 and George #4?

This difference does not preclude George #4 being the son of Henry #1. Mutations happen randomly, and George could have passed this mutation down through his Y line to Ira #43.

• d) Marker DYS449=30 for Louis Jr. #44 but DYS449=31 for the other tested men. What does this tell us about where this change may have occured?

In Ira Smith #20, Louis Smith #32 or Louis Smith Jr #44.

• e) Marker CDY=35-38 for Robert #33 but CDY=35-39 for the other tested men. What does this tell us about where this change may have occured?

In either Perry Smith #21 or Robert #33.

• f) When all the differences are counted, there are three markers that vary between these men (DYS464, DYS449 and CDY). And a fourth difference between these men and others who are in the same group in the Smith SP. When four differences are found, test-takers are "distantly related" while three differences would indicate they are related. Are there other factors to incorporate into the analysis?

There is strong documentary evidence for each of these family lines, and in "fast-moving" markers, the changes seen in the STR results from each of the male lines, as well as another differentiating #1 Henry Smith's male descendants from the other Smiths in the Smith Surname Project. While 4 is a larger difference than we would expect, the fact that they are all "fast-moving" markers makes it easier to account for.

• g) Could additional DNA testing supply more evidence to apply to this research question?

Testing descendants of the "other" George Smith could be useful. Also, since five men have already given test samples, some or all of them could upgrade their tests to autosomal DNA -- and Big Y if possible.

7. All of the living cousins agree to take an atDNA test. The table below shows some of the results in the match list of Ira Smith #43.

Cousins match list to Ira Smith #43

• a) Is this a reasonably sized group of test takers to possibly lead to credible conclusions about relationships?

Yes, each line is covered.

• b) What might be a first step in analysis?

Do they have family stories about #1 Henry? More documentation, photos, memories? Are these folks all descendants of Henry #1 and unnamed #2?

First level of credibility: common descendancy of a large enough group of testers.

• c) Should we be concerned about the depth and accuracy of the test-takers' trees?

No. We have them in the tree already. It would be good to share the tree with them so they can make any corrections necessary though. 

Second level of credibility: reliability of the family tree.

• d) Using the information in the table above, what might be a next step in our analysis?

One could enter the match cMs into the Shared cM Tool at https://dnapainter.com/tools/sharedcmv4 and see the relationships "on paper" match the DNA cM results. If any are unexpectedly large, could there be other shared ancestors?

Third level of credibility: consistent DNA match results.

The table below is a matrix indicating which cousins show as "In common with" each other.

Which cousins show "In Common With" each other

• e) Using the information in the table, what might be the next step in our analysis?

Check with each tester to see if the other cousins are shown by the testing company as matches.

Fourth level of credibility: cousins "on paper" all matching with one another. 

Selected shared segment details; page 122

• f) Using the information in the table above, what might be the next step in our analysis?

Fifth level of credibility: match segment data.

Since so many lines are covered, it might be possible to pinpoint from which ancestors certain segments were inherited. This requires segment data and either a tool like GDAT or DNA Painter, or a spreadsheet.

Triangulating by DNA segments is the gold standard when used with excellent relationship evidence in a proof argument.