Acceptance Critieria as a function of 15 Generation mutation events

Return to The Tapestry Homepage		Return to Methods
Contents 1 Background 2 Results 3 Interpretation	This is one of a series of articles on Genealogical Methods, prepared in association with The Tapestry. See Index for a list of related articles.

Background

A 37 marker "seed" haplotype to represent a Patriarch ancestor. This seed was used to computing haplotypes for 100K descendants at 15 generation level. (i.e., 1500K mutational events were captured.) Changes in descendant haplotpes were based on the probability of a marker specific mutation in each of the 1500K descendants, using Source:Chandler, 2006 as the source for the mutation rates. "Runs" were made at varying marker levels up to the 37 markers for which Chandler provides mutation rates. Haplotypes were compared to each other based on the change in marker values, expressed as a percentage of the total number of markers. The objective was to determine the "similarity" that would have to be accepted to ensure that 95% of the 15th generation descendants were captured.

Results

Interpretation

The acceptance criteria necessary to achieve ∂95 decreased with increasing number of markers included in the analysis. Results (shown above) for the first 37 markers for which mutation rates are given by Chandler show that ∂95 decreases with the number of markers included, in a simple power function. This allows values of ∂95 to be projected for larger numbers of markers that might be tested, including in particular, in the 67 and 111 marker tests in the FTDNA system.

Increaseing the number of markers tested can substantially decreases the ∂95 acceptance criteria value. However, there is a diminishing return as more and more markers are added to the test. The most dramatic improvement comes when marker levels increase above about 25 markers. While there is progressive improvement after that point, the improvement steadily diminishes.