Diana wrote, "They {FTDNA} do test them {19b, 464e and 464f}. However, as the vast majority of people don't have those markers, FTDNA simply doesn't include them in the count for advertising purposes." I recently had the opportunity to learn more about the curious DYS 464, to answer a project member's query. According to FTDNA's FAQ, 464 is a marker (locus?) that replicates 4 to 6 times with the replications designated 464a, 464b, 464c, 464d, 464e & 464f. Only about 1.5% of all those tested have more than four replications of 464, i.e., 464e &/or 464f. More curiously, FTDNA reports from low counts to high, that is, "464a" is always the lowest count of the 464 series, 464b the next lowest, and so on. Mismatches on 464a-d may be due to reordering of the results, rather than disagreements on specific loci. In our member's case, there were two mismatches in 464a-d, but they could not be explained by re-ordering; no order would have produced fewer mismatches. As these are highly volatile, I suspected (& FTDNA confirmed) that this was the interpretive equivalent of slightly more than one mismatch on less volatile markers. -rt_/)

> -----Original Message----- > From: y-dna-projects-bounces@rootsweb.com on Behalf Of Ralph Taylor > Sent: Monday, June 14, 2010 7:48 PM > To: y-dna-projects@rootsweb.com > Subject: [Y-DNA-projects] Medley Family DNA & DYS 464 > <snip> > > I recently had the opportunity to learn more about the > curious DYS 464, to answer a project member's query. > According to FTDNA's FAQ, 464 is a marker (locus?) that > replicates 4 to 6 times with the replications designated > 464a, 464b, 464c, 464d, 464e & 464f. Only about 1.5% > of all those tested have more than four replications of > 464, i.e., 464e &/or 464f. I think I would describe it as there being 4 to 6 *copies* of DYS464, which is why it and others markers like it are called "multi-copy" markers. Replication is the process of a chromosome making a copy of itself, and all the markers replicate. Other multi-copy markers are DYS385a/b, DYS459a/b, and YCAIIa/b. > More curiously, FTDNA reports from low counts to high, that > is, "464a" is always the lowest count of the 464 series, 464b > the next lowest, and so on. Mismatches on 464a-d may be > due to reordering of the results, rather than disagreements > on specific loci. With most multi-copy markers, it is not possible to tell the actual order of the alleles, so, by convention, they are reported lo-hi. I make it a practice to reorder the alleles to produce the fewest difference possible when calculating GD (genetic distance). I don't actually re-order them in my data table because that is not how FTDNA reports them, so I don't want to confuse people, but I reorder them for the process of determining GD (genetic distance). For example... I have a case where two family members have these values at CDYa/b, keeping in mind this is very volatile marker prone to multi-step mutations. Their results were reported: 41 42 40 41 Taken as a straight count, this is a GD of two, one at CDYa and one at CDYb. However, if you re-order them: 41 42 41 40 this can be read as a single two-step mutation at CDYb for a GD of only 1 (one mutation event). I have another case with 21 family members tested, where at DYS464, 20 of them are 12 14 15 15 However, one member of the family has six alleles at that locus. If we line them up as reported and read them literally, we get a GD of 4, for the four differences: 12 14 15 15 12 12 14 15 15 15 However, what probably really happened is that the 12 allele and one of the 15 alleles were duplicated in a mutation single event, which is apparent if we reorder the alleles, which then brings the GD down to only 1, which is a more realistic distance within the family: 12 14 15 15 12 14 15 15 15 12 It is possible to determine the order of the 385a/b alleles with a Kittler test. Most R1b1b2's will actually turn out to be hi-lo with a Kittler test. > In our member's case, there were two mismatches in 464a-d, > but they could not be explained by re-ordering; no order would > have produced fewer mismatches. As these are highly volatile, > I suspected (& FTDNA confirmed) that this was the interpretive > equivalent of slightly more than one mismatch on less volatile > markers. Whether you lean towards maximizing the GD or minimizing it depends in part on how closely the individuals match on other markers. If they are otherwise a close match, it's more probable that a difference of two or three at one marker happened in a single mutation event. If the individuals are not otherwise particularly close, the difference could be the result of several single-step mutation events. There is a way to determine whether big changes in volatile, multi-copy markers are the result of several single-step mutations or one multi-step mutation and that is by testing cousins. As you test more and more distant cousins, you can watch for the appearance of the mutation(s), which will tell you in whom it occurred and in how many stages. Diana