What is the probability that an 111 marker YDNA signature could remain intact for 10 generations?

Question

The issue at hand is related to an NPE that may have occurred at about generation 5, but some argue it took place before generation 10.  Hence the question.

It's possible, I suppose, but it does not seem very likely.  I would think that over 10 generations at least one of those fast moving markers would change.

Comments

Here are some resources on mutation rates of Y DNA from the International Society of Genetic Genealogy. https://isogg.org/wiki/Mutation_rates

Answer 1

If you use FaceBook there are several y-dna related groups that have people who can answer that question.   
Alexa found this article at National Library of Medicine
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1181912/
It would appear that 5 generations would be more likely than 10. - the exact odds are unknown.

Answer 2

Can you use FTDNA's TiP tool? That would be able to tell you that there is an X% chance that the MRCA was within 9 generations. The probability that the MRCA is 10 or more generations back would then be 100-X.

There would no doubt only be a small percent chance that the MRCA was 10 or more generations back if the test takers had an exact match at 111 markers. I have a genetic distance of 6 from another test taker at 111 markers. Even with that gd of 6, TiP says that there is a 71% chance that our MRCA is within 9 generations. I suspect a gd of 0 would show a probability in the upper 90s and yield a correspondingly very low probability that the MRCA was 10 generations or more back.

Comments

A bit of background on the TiP tool: FTDNA solicited customer submissions of deep-rooted pedigrees where several people had been tested. In principle, that is an excellent approach: you can get more data than you can with father-son studies. In practice, it appears that there was some bias in submissions. People who had pedigrees with no mutations didn't think their results would be interesting, so they didn't bother to submit their results. That means that FTDNA used a higher mutation rate, which in turn means that the generation estimates are over-optimistic.

---

"A bit of background on the TiP tool: FTDNA solicited customer submissions of deep-rooted pedigrees where several people had been tested. In principle, that is an excellent approach: you can get more data than you can with father-son studies. In practice, it appears that there was some bias in submissions. People who had pedigrees with no mutations didn't think their results would be interesting, so they didn't bother to submit their results."

Ann - Do you have a source for this? From a brief google search, I haven't been able to find anything online

---

I guess you could say the "source" is my memory from the early days of the GENEALOGY-DNA mailing list on RootsWeb, but this is the closest I could find:

https://web.archive.org/web/20080217040013/https://jogg.info/32/editorial.htm

JoGG articles have new URLs, but I didn't take the time to locate those.

---

Thank you all for each and every answer received.  If there's a consensus here, it is "Very unlikely".  I will follow up on the various suggestions about resources related to this question.

Bill

Answer 3

The probability is extremely low, so low as to not be possible actually...  SNPs are the only mutation markers that do not change over time. STRs will move around over time as the number of mutations at each STR marker will move up or down within a cycle over time. 10 generations can be as little as 250 years (25 years/generation) or as much as double that in some family lines. So your question implies a duration of as much as 250 to 500 years ago to present.

Answer 4

Less than one percent (i.e .311%) according to https://www.mitoydna.org/Public/MutationCalc if 20 transmission events.

Comments

Oh, my -- I wrote that calculator 20 years ago! One critical parameter is the mutation rate. I don't know if that's been updated for the 111-marker test. Anyone?

Answer 5

I'm not sure an "exact" could ever be anticipated. FTDNA predicts our group of 10 y participants will match 99-100% within 20-24 generations. Yet with our researched paper trails we are connected to 2 brothers in 1550 England. We have 2 tested to 111 markers and though when their tests were entered they were both at 67 markers distance of 2, now they are 111 matches at distance of 3. They are 5C1r and connect to the same ancestor, Capt William Mitchell, most of us connect to his brother Thomas but as a group we are matched. Furthermore, we really don't see any need for additional testing of our group to 111 markers, or additional snps. We would all predictably be same results. We are FGC18441, the generational differences will vary but for my brother it is gen 12.

Comments

Well you could test further - to a BigY or other NGS. Those results would identify the paternal pedigree with definitive Y-markers. And not the calculated probability that these Mitchell men are related through kinship, esp if the paper record says as much. The certainty of FGC18441 should be validated by the right type of test, and not inferred.

---

I respecify disagree. My experience is that we all follow as a group. No add testing is needed.

Answer 6

Alot of the answer to that would depend on the markers that make up the "signature". Each STRs (at least tracked up to y67) appears to have different rates of mutation. Even that data is over a decade old now, last I checked.
One you examined all that you would have a more informed idea of the possibilities.
One of the problems arises in an STR marker that might mutate one in a 1000 years. Where does the 1000 years start and end? That actually would apply to any of the STRs and their mutation rates.
If the family tree is extensive enough, you might be able to develop an idea of that, if there are differences. You are saying there are not differnces in this case.
The point is, if the signature is made of of STRs that mutate at a snails pace, that would say one thing... If they are made up of STRs that mutate every generation or every other generation, that would say another... If you had a mixture of the tow, you would need to consider forward and then back mutations and the like, etc etc.
In other words, there is not an easy answer. :)

Answer 7

Yes.  So far I have found two DNA cousins who's direct lines departed mine around the 9 and 10 generation mark. And one who predates that time period that we linked with Big Y.   We verified the links via paperwork and then the Y-DNA on family tree.  Previous to this I was just guessing and I only had solid links back to my 7th great grand father who was born in 1678. 

Comments

Very interesting!  Can you provide Kit numbers and the surname involved?