GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2005-07 > 1120278508


From: "Ken Nordtvedt" <>
Subject: DYS455= 8
Date: Fri, 1 Jul 2005 22:28:28 -0600


Lots of debate today about DYS455=8 and I1a. Some hard numbers might help.

Putting into Sorenson database the 6 most invariant marker modal repeat values found in the alleged I1a population other than DYS455, I could then search for the distribution of DYS455 repeat values. Here are the counts after expunging multiple family contributions from the small populations (I used 426 = 454 = 11; 438 = 10, 388 = 14, 459a,b = 8,9; YCAIIa,b = 19,21 and probably lost 10 percent of my DYS455 = 8 population by so restricting things; with much work I could recover those)

DYS455 = 7 (0)
DYS455 = 8 (892)
DYS455 = 9 (4)
DYS455 = 10 (0)
DYS455 = 11 (2)

And DYS455 = 11 is the repeat value for just about the entire rest of the European population --- R1b, R1a, I1c, N, G, ..... Probably leakage of 2 haplotypes from this entire population has occured. This discrimination is achieved with 7 marker haplotypes, albeit a selected 7 markers.

I have not done it, but the 4 DYS455 = 9 haplotypes could be checked at the usual suspects, DYS19,390,385a,b, and probably found to be "bonified" I1a-like haplotypes whose 455 has mutated from 8 to 9 over the 10,000 year course of their ancestral lines.

Rootsi has measured the I1a SNP M253+ for a couple hundred haplotypes from around Europe. But they were short haplotypes which did not include DYS455.

Maybe just a single handful of haplotypes in the world which include DYS455 have actually been SNP tested for M253 or P30 so far.

Everything learned about the population of DYS455 = 8 haplotypes had nothing to do with SNPs. The population could just as well be called the "DYS455 = 8 group"

Now admittedly the accidental isolation of 8 repeats at 455 for this group and the extremely slow mutation rate of 455 made this possible; so other haplotype populations one wants to study can not be defined or described by a single marker but require definition in terms of several key marker repeat values. But the principle is the same; most everything about the populations including their geography can be studied without reference to any SNPs. If these objective populations can be connected to a SNP all the better. That makes relating these populations to others in a time-ordered tree (via the tree of the SNPs) that much easier, and makes other studies of the populations more efficient.


This thread: