How many generations removed from a DNA test taker assume that the DNA will be significant enough to identify ethnicity?

+12 votes
500 views
My family research on-line and direct searches over the years suggest that one of my 4th. paternal g.grandmother's name is Sytha (possibly Lucas) Lester. Some of this research, especially on Ancestry, suggests that Sytha's ethnicity is Cherokee Indian.  My recent autosomal test states that I have no Indian DNA. 4th. generation g.grandparent probable percent DNA would be roughly 0.15625. What should I assume from this autosomal test, if anything?
asked in Genealogy Help by Denny Lester G2G Crew (970 points)
retagged by Abby Glann
Since we're only going back 160 years, "Cherokee" might mean only part-Cherokee.
Denny, this is a little off-topic, but is Sytha a known native-American name?
Edie, Thanks for responding to my previous comments on WikiTree. The short answer is "No."  Not that I have been able to establish in over 25 years of research. I have found "Sytha", as mentioned in my other comments, in a few places that I searched on-line, and none were Cherokee. Official Cherokee records listed on-line - marriages, census records, court records and births - do not go back that far, so I have not been able to establish Sytha, possibly "Lucas" Lester's ethnicity. FYI. the name has been used down through my family and many collateral lines of Lesters.

4 Answers

+11 votes
I agree with your figure but DNA is not taken equally from each parent. It's perfectly possible for you to have less than 0.05% from that particular ancestor. You could easily miss having the markers that have been identified as deriving from Cherokees. Just googling for Cherokee DNA reveals there are in any case many issues with this sort of testing so I wouldn't give it much weight.
answered by Matthew Fletcher G2G6 Mach 6 (66.7k points)
Thank you for your reply. I assumed as much concerning the probabilities of any DNA segments after so many generations, but her unique name, "Sytha" and various spellings was passed down to my grandfather's generation in many collateral Lester families. Using that name to that extent is indicative of a special person that I wanted to find out more about. But about all I can definitely prove is her given name as the wife of my 4th. g.grandfather and that she remarried after his death in 1787 to a Ballinger/Ballinger in Pittsylvania Co., Virginia and was living at the time of his death in 1821 according to his will.
+9 votes
If you are talking about a great-great-great-great-grandmother, it is not at all surprising that you do not have any DNA of her's.

This is one of the key points that Blaine Bettinger makes when he discusses how people have genealogical family trees that are based on family stories and records and genetic family trees that are based on the DNA evidence that you are using.  In autosomal DNA, ancestors begin "falling off" of a person's genetic family tree 5-9 generations back.  None of the genes from your 4X-great grandmother were passed all the way down to you:

http://thegeneticgenealogist.com/2009/11/10/qa-everyone-has-two-family-trees-a-genealogical-tree-and-a-genetic-tree/
answered by Ray Jones G2G6 Pilot (149k points)
Thank you. This attached article reminded me of the old biological principle in genetics:  The Law of Independent Assortment.  I still do not have an answer to my original question regarding ethnicity, but I think I now understand the results of my autosomal test.
So true, just looking at my grandchildren's dna variation inherited from each grandparent is amazing. from 22% to 27%. Additionally, since only about 1% of dna shows up in consumer tests, I wonder how many bits of information are there somewhere, but we are not seeing the whole picture.
+4 votes
GEDmatch's Eurogenes K=13 indicates that 3.9% of your chromosome 1 is Native American.  It is my understanding that above about 4% is considered real (not just noise).

Sincerely, Peter
answered by Peter Roberts G2G6 Pilot (442k points)
Peter, I checked your link on GEDmatch and reentered my auDNA results and selected Eurogenes K-13 and see where you got the 3.9% on chromosome 1. Looking at other data I have that does not indicate American Indian ethnicity, aren't we just looking at the way different people are setting different parameters analyzing the same raw DNA? It appears to me that since every nucleotide on each chromosome in our DNA is not identified in any one DNA test, it just depends on what segments on each chromosome the test makers used in their analysis which still leaves a lot of "unknowns" and continued guessing on our part. Any comments accepted....
+3 votes
Only about 1% of our dna is tested, which makes me wonder if when we are able to easily do whole genome dna testing, we'll find more markers for our ancestry that are hidden from direct to consumer tests.
answered by Elizabeth Blake G2G Crew (920 points)
I think you are absolutely correct, and even with an autosomal test that records maternal and paternal lines, the number of generations removed makes it very unlikely that I would have any DNA from that generation. But if you have other family members that take current available DNA tests, small segments of nucleotides (part of a chromosome) may show up in a sibling and or cousin. Even with a more thorough testing process of the entire genome, biological diversity is insured due by the process of the formation of a sperm and egg (maternal and paternal). Thus siblings are genetically slightly different for a reason and any clue of ethnicity could bypass me entirely after that many generations.  Neither the mitochondrial or yDNA tests would produce what I was searching for with respect to her ethnicity in my paternal line.

Related questions

+2 votes
0 answers
+8 votes
1 answer
155 views asked Jun 28, 2015 in Genealogy Help by Curt Cornell G2G Crew (370 points)
+10 votes
2 answers
+2 votes
3 answers
+5 votes
2 answers
+9 votes
4 answers
+1 vote
2 answers

WikiTree  ~  About  ~  Help Help  ~  Search Person Search  ~  Surname:

disclaimer - terms - copyright

...