Understanding genetic distances
One of the joys of using DNA as a tool for genealogy is having multiple tests available to see relationships in different ways.
One of the downsides of using DNA as a tool for genealogy is having multiple tests available to see relationships in different ways … and not seeing them work together nicely the way we expect them to.
Case in point: the YDNA – autosomal mismatch.
Reader Margaret Mitchell had a male cousin take a DNA test for her, testing both his YDNA and his autosomal DNA. YDNA, remember, is the kind of DNA that only men have and that’s passed down through the male line from father to son to son (and so on) largely unchanged through the generations.1 And autosomal DNA is the kind of DNA we all inherit from both of our parents and that’s used, in genealogy, primarily to find cousins back to roughly the fourth to sixth cousin level to work with collaboratively.2 The tests were done at Family Tree DNA, the only major DNA testing company currently offering both YDNA and autosomal DNA tests, the latter called Family Finder at that company.
When the results came in, one of the matches was particularly interesting. Both men had been tested to the level of 111 markers for YDNA and both had taken the Family Finder test. They were shown with a genetic distance of four (GD-4) on the YDNA test — and didn’t show up as a match at all on the autosomal test.
And that left Margaret confused: “if he matches at a 4 genetic distance with another Y-111 DNA tester who has also taken the Family Finder DNA test,” she asks, “why wouldn’t I find this person on my cousin’s Family Finder as a 4-6 generation Cousin match?”
The short answer is, because that GD-4 result — genetic distance of four — doesn’t mean they’re four generations apart.
The longer answer begins with a closer look at the concept of genetic distance in YDNA results. It’s not a direct measure of how many generations back the common ancestor may be but, rather, is “the number of differences, or mutations, between two sets of results. A genetic distance of zero means there are no differences in the results being compared against one another, i.e., an exact match.”3
So we need to consider not just how many differences there are, but consider those difference in light of what we know about how often differences occur when YDNA is passed from generation to generation. In other words, the question could be posed this way: “If two men share a surname, how should the genetic distance at 111 Y-chromosome STR markers be interpreted?”
Which is exactly how Family Tree DNA phrases the question in its Learning Center. And it offers these answers for those who’ve tested at 111 markers:
• A genetic distance of zero (GD-0), suggests that two men would be “very tightly related”: “A 111/111 match indicates a very close or immediate relationship. Most exact matches are 3rd cousins or closer, and over half are related within two generations (1st cousins).” The odds are 50-50 that the common ancestor will be within two generations, 90% in four generations, 95% in five generations and 99% in six.4
• A genetic distance of one (GD-1), suggests that two men would be “tightly related”: “A 110/111 match indicates a close relationship. Most one-off matches are 5th or more recent cousins, and over half are 2nd cousins or closer.” The odds are 50-50 that the common ancestor will be within three generations, 90% in six generations, 95% in seven generations and 99% in nine.5
• A genetic distance of two (GD-2), also suggests that two men would be “tightly related”: “A 109/111 match indicates a close relationship. Most matches are 7th cousins or closer, and over half are 4th or more recent cousins.” The odds are 50-50 that the common ancestor will be within five generations, 90% in eight generations, 95% in nine generations and 99% in 11.6
• A genetic distance of three (GD-3), suggests that two men would be “related”: “A 108/111 match indicates a genealogical relationship. Most matches at this level are related as 9th cousins or closer, and over half will be 5th or more recent cousins. This is well within the range of traditional genealogy.” The odds are 50-50 that the common ancestor will be within six generations, 90% in 10 generations, 95% in 11 generations and 99% in 14.7
• A genetic distance of four (GD-4), still suggests that two men would be “related”: “A 107/111 match indicates a genealogical relationship. Most matches at this level are related as 10th or more recent cousins, and over half will be 6th or more recent cousins. This is well within the range of traditional genealogy.” The odds are 50-50 that the common ancestor will be within seven generations, 90% in 11 generations, 95% in 13 generations and 99% in 16.8
So that GD-4 for these two men doesn’t mean they’re within four generations or even fourth cousins at all. They could be 10th cousins — or even more distantly related — and their YDNA simply didn’t change much as it passed down generation to generation from a common ancestor very far back in time.
Now it’s possible to find all this out on the Family Tree DNA website if you know where to look. But there’s also an easy double check right in the YDNA results: it’s called the TiP report — for Time Predictor9 — and it’s one of the icons under the names of all of the matches for anyone who’s YDNA-tested there, the orange box with the letters “TiP” inside.
What this tool does is calculate the likelihood of finding the most recent common ancestor within a number of generations, considering which markers are different and how quickly or slowly those particular markers are known to change. It’s a probability estimate, not a guarantee, and is designed simply to help us narrow down the best place to start looking for a common ancestor between two testers.
The default setting is to assume that the two men don’t share a common father (they’re known not to be related within one generation) and then present the odds every four generations. The results can be refined: if you know they don’t share a common grandfather or great grandfather, for example, you can change the dropdown box to reflect that, and you can see the results calculated for every generation rather than every four generations.
For one of my relatives, for example, the default settings suggest that we might find the common ancestor for a GD-4 match within four generations roughly 43% of the time and within eight generations roughly 88% of the time. But we know that one of his GD-4 matches does not share a common ancestor within three generations (no common father, grandfather or great grandfather is possible). The TiP tool tells us, then, that the odds of finding the common ancestor drop to roughly 35% within four generations and roughly 86% within eight.
Bottom line: “genetic distance” isn’t the same thing as “number of generations to the common ancestor.” So it’s often the case that an apparently close YDNA match may be 10 or more generations back, and so won’t end up being in our autosomal match list of fourth to sixth cousins at all. And using the TiP report will quickly give us an overview as to why not.
Cite/link to this post: Judy G. Russell, “A tip on TiP,” The Legal Genealogist (https://www.legalgenealogist.com/blog : posted 3 May 2020).
- ISOGG Wiki (https://www.isogg.org/wiki), “Y chromosome DNA test,” rev. 4 Sep 2019. ↩
- Ibid., “Autosomal DNA,” rev. 13 Aug 2019. ↩
- FamilyTreeDNA Learning Center, Glossary, “Genetic Distance” (https://learn.familytreedna.com/ : accessed 3 May 2020). ↩
- Ibid., Expert’s Handbook: Y-DNA Testing : Y-DNA STR Testing, “If two men share a surname, how should the genetic distance at 111 Y-chromosome STR markers be interpreted?” ↩
- Ibid. ↩
- Ibid. ↩
- Ibid. ↩
- Ibid. ↩
- ISOGG Wiki (https://www.isogg.org/wiki), “TiP,” rev. 13 Dec 2014. ↩
Good explanation of Genetic Distance. I still find the TIP tool to be a little confusing but I am working on that. Is there anything similar for mtDNA? I have 67 perfect matches and no connection to any of them that I can find. I know the female lines change every generation but it seems I should find at least one common ancestor out of 67.
No, nothing similar for mtDNA because it’s much much more stable than YDNA and changes far less than YDNA. It’s considered to be entirely possible for two people to be perfect mtDNA matches — and still have their common ancestor so far back in time as to be impossible to determine.
Thank you so much for your explanation of genetic differences! I have a much better understanding of this now.
I thought by having my 1st Cousin take the Y-111 DNA test would certainly put me on the path of knowing our paternal line of ancestors. Unfortunately it has not given me concrete evidence of who they are.
The Y-DNA results with haplogroup R-296>L-513 has given me a glimpse of where our ancestors originated from thousands of years ago, but I was in hopes to find something more recent.
I have been at a road block for 20+ years with my Stewart family.
The furthest I’ve been able to get is with our 3rd Great Grandfather Samuel Stewart born 1779 in Montgomery, Maryland. Died 1832 in Georgetown, Washington, D.C..
No clue who his parents were and no signs of siblings.
I did have my male Cousin take a SNP test for the Stewart Royalty bloodline, but that tested negative.
Any suggestions about which direction I should go now would be much appreciated!
BTW, I had the pleasure in attending one of your DNA lectures down in St. Mary’s County, MD a few years ago. You were fabulous!
Thanks for the kind words. I’m told by people who understand YDNA testing better than I ever will that having BigY-700 tests done in a big surname project (and I’m guessing that an R Stewart haplogroup is gonna be a lot of people) can help narrow the group down enormously. Skip Duett will be speaking on this very issue in the virtual NGS conference recordings, so it might be worth making sure you register for one of the packages for that conference.