-----Original Message----- From: Terry In your method, you require some estimate of the variance of that theoretical allele distribution, and the way you are computing that is by assuming that variance is the same as the variance of the alleles (or repeat values) that you happen to see in the surviving population. [[You apparently don't understand what is done. The variance of the allele distribution is not assumed or used for purposes of the tmrca estimate; that estimate comes from the observed haplotype differences in hand. You would do better asking what I do, rather than inventing your false notion of what is done. The theoretical allele distribution plays no part in the tmrca estimate. It only plays a part in determining the sigma or confidence interval associated with a tmrca estimate. Given the tree leading to the sample of haplotypes, what would be the distribution of variance outcomes if nature had had the opportunity to throw its str mutations into that tree many times? Our sample haplotypes are nature's ONE example of doing this; so the variance nature gives us in this single case is one realization from that theoretical distribution --- all the theoretical distribution can tell us is the statistical range of typical error "off the mark". KN ]] That is the problem. The variance you see in the surviving population is tainted by the population history, and especially so since many male lines go extinct, and only a few initial male lines get to survive. [[ What happens on branch lines that went extinct, or on branch line segments that do not lead to the sample population of haplotypes is all irrelevant to estimating the age of the pruned tree which leads to the sample of haplotypes. What actually happened which led to the sample of haplotypes had to happen on the branch segments which lead to the sample of haplotypes --- no where else. KN]] [[The detailed demographics of the tree will affect the expected value of the coalescence age estimate, but not the interclade node estimate. This has been stressed for years, and that's why one must call coalescence age an estimate of an abstraction rather than a specific historical event.]] The Walsh approach is less affected by details of the surviving population. The method also gives the probability distribution for the TMRCA - so one can put say a 95% confidence interval on the TMRCA estimate. But to do that one needs to assume perfect confidence in the STR mutation rates, and the mutation process, which I don't think we can assume or know just yet. [[ And what in a formula nutshell is this "Walsh approach"? I bet you it is one of the variance or GD methods some of us use all the time. The only thing I did not include above is the tmrca estimate using variance from assumed founding haplotype. But that is not recommended for the obvious reason; we have to infer the founding haplotype which can be subject to error. Yes; as the age estimates go deeper into the past, the true nature of the mutation rules become increasingly important, especially for the fast strs which could very well mutate more than once on any branch line of descent in the tree under consideration. This also has been discussed for some time. ]] Terry ------------------------------- To unsubscribe from the list, please send an email to Y-DNA-HAPLOGROUP-I-request@rootsweb.com with the word 'unsubscribe' without the quotes in the subject and the body of the message