Chat
|
Daily Search
|
My GenForum
|
Community Standards
|
Terms of Service
|
|
|
Hi from Mt
Yeah, I agree with almost everything you state. Using only Y chromosome just does not give enough info to make many of the correlations that are being made. In addition, any statistical analysis bends the data. Example: kit #8179 from UK has 5 mutations in a 67 site test,having 3 mutations in the first 37 sites. It is accepted that 3/37 is considered not related within a reasonably measurable time frame. The analysis used by FTdna overrides this view, and they are able to say there is a UK connection between group 3a and the UK test. No proof, very slim probabilty of a measurable connection.
You are right, using the infinite allele model to determine TMRCA is more uncertain in either a plus or minus, so it is a leap to claim a relationship between group 3a and 3b. It also a leap to say there is no relation using that analysis. If the single allele model is used to determine TMRCA the uncertainty is less that 3a and 3b are not related. The probability is that 3a and 3b are not related. Is it proof of yeah or nay of relationship? Nope Just not enough info on the Y chromosome.
It is my understanding that there is no documentation re the origins of families that the dna tests show mutation on site 6 but no mutations on site 21. If one accepts dna evidence, one can not dilute the evidence one way or another by injecting naming patterns or oral histories. It seems naming is influenced by popularity in time and place. Undocumented histories are influenced by everything. Therefore, using dna only these people more probably belong in group 3a than 3b. Mathematical probability is much more likely that a single mutation occurred on site 6 than a double reverse mutation occurred on site 21. Again, not enough info on Y.
To hang your hat on any relationships using only Y chromosome dna testing is a leap of faith in any direction. The tests are out there that increase certainties of relationships but currently too expensive and unavailable. In the meantime we must realize that there are lies, damn lies, and statistical analysis.
Notify Administrator about this message?
Followups:
|
| http://genforum.genealogy.com/pace/messages/4159.html | ||
 |
Link to GenForum Add Forum |
|