SNP-Based Calculations of Time to MRCA for R1a-Y2619 Ashkenazi Levites

[This page, first posted on this website in 2014, is posted here as a matter of historical interest. The Big Y mutation rates refer to mutation rates for SNPs reported by the Big Y-500 test; mutation rates for SNPs reported by the Big Y-700 test are comparable to those discussed below for Full Genomes Corp. testing.]

Because SNPs mutate at a regular rate, it is possible to use full Y-DNA SNP results to date the time at which subclades branched off from one another. Michał Milewski has estimated in a post on the Eng.Molgen website that based upon his preliminary analysis, on average, a mutation may occur once every 88 years on the SNPs reported by Full Genomes Corp., and once every 150 years on the SNPs reported by FTDNA's Big Y test; he cautions that the estimated time to an MRCA may not be very reliable when comparing individual Big Y results, where the analysis could be more easily skewed by random mutations of SNPs.

Milewski has analyzed the time to an MRCA for various subclades of R1a1a Ashkenazi Levites, based upon the R1a1a Ashkenazi Levites and closely related men for whom we have SNP results as of August 1, 2014. He has calculated the following time to an MRCA for the following clades and subclades, separately estimating the time to an MRCA based upon: (1) all Y-DNA SNPs (i.e., the results from the men who have tested through Full Genomes Corp. or who have had full Y-DNA testing for scientific purposes); and (2) all Y-DNA SNPs that are tested by the Big Y test (i.e., the results from the men who have tested through the Big Y test or Full Genomes Corp. or who have had full Y-DNA testing for scientific purposes).

Milewski calculates the following (the number of mutations refers to the average number of mutations downstream from the identified SNP level) (note that the Big Y-based calculations have been refined since the version of the web page referenced in Anatole Klyosov's comment on these calculations):

Subclade Time to MRCA (all Y-DNA SNPs) Time to MRCA (Big Y SNPs)

F1345 4,338 years (49.3 mutations) 4,590 years (30.6 mutations)

CTS6 3,282 years (37.3 mutations) 3,090 years (20.6 mutations)

Y2619 1,346 years (15.3 mutations) 1,590 years (10.6 mutations)

Y2630 1,056 years (12 mutations) 1,238 years (8.25 mutations)

YP265/YP264 616 years (7 mutations) 705 years (4.67 mutations)

As discussed here, pending the receipt of additional SNP results, it currently appears that: (1) Y2619 is characteristic of R1a1a Ashkenazi Levites; (2) Y2630 may be characteristic of R1a1a Ashkenazi Levites who are close to the mode; and (3) YP265/YP264 may be characteristic of descendants of the founder of the Horowitz rabbinical family, who moved to Horovice, near Prague, in the 1470s. [It was subsequently determined that YP268, downstream from YP265/YP264, defines the subcluster including members of the Horowitz rabbinical family and their closest matches.]

Milewski comments:

* * *

Honestly speaking, I don't find anything unexpected in those two sets of estimates. For example, I don't know anything that would suggest that F1345 and CTS6 are significantly younger than 4,000 and 3,000 years, respectively. On the contrary, since we know only nine Z93-specific SNPs upstream of CTS6 (including F1345, F2997, Z2122, Z2121, Z2124, Z94, Z95, Z2479 and Z93) and Z645 is commonly considered to be about 5,800 years old, this would suggest that F1345 is likely to be about 4,500 to 5,000 years old, or at least older than 4,000 years.

The age of the Ashkenazi Levite clade (or Y2619), as calculated based on the FGC data, is indeed slightly older ([1,590] years) than expected based on the previously known STR-based estimates. However, I would still consider it acceptable. Importantly, testing some members of the Y2630* group and comparing the number of their downstream SNPs with that found in the YP264 members will be crucial for refining the above estimates, as this may either confirm these higher TMRCA values or make the age of Y2630 look much younger (in case the number of private SNPs in those additional sublineages of Y2630 turns out to be significantly lower).

The same can be said about my Big Y-based estimates, as it seems that the number of private mutations in kit no. 65218 is exceptionally high, so testing more people should give us a much more secure (and probably slighly lower) estimates for both Y2630 and YP264. [The Big Y results for kit no. 248865 and FGC results for kit no. 241703, subsequently received, confirmed Milewski's belief that the number of private mutations in kit no. 65218 are high and that Big Y-based estimates based upon a larger sample size yield a shorter TMRCA for both Y2630 and YP264.]

* * *

For discussions of STR-based calculations of the age of the R1a1a Ashkenazi Levite cluster, click here.