Archiver > Y-DNA-HAPLOGROUP-I > 2009-03 > 1237475532

From: "Ken Nordtvedt" <>
Subject: [yDNAhgI] SNP Searches and Tree Time Depth
Date: Thu, 19 Mar 2009 09:12:12 -0600

So let me summarize what I found about an I2b1 region search for a new snp and an I1-AS region search --- by paid-for examination of 100,000 or so nucleotide sites of the Y. Chances of success are much about TIME DEPTH of the Tree region you are searching in with your compared dna samples. Fortunately, the time depth of the I2b1 tree seems like about 10,000 years, while the I1-AS tree time depth seems about only 5000 years. These depths are measured back to the oldest node the branch line passes through as it goes from one dna donor in the tree to another.

Total chances of finding a new snp comparing a I2b1-Roots to an I2b1-Continental-1or2 are high, close to 1. But chances the snp will not be private but widely useful for hobbyists are maybe about .73.

Total chances of finding a new snp comparing an I1-AS1 with an I1-AS3 are perhaps .75. Chances a widely useful (non-private) snp will be found for I1-AS in that 2-donor comparison are perhaps about half that.

A search program for a widely useful I2* snp will probably have good numbers; I2* tree depth is quite large.

A search program for a useful I2a sector snp will have large probability of success, given the time depth of up to 20,000 years for that portion of the haplogroup I tree. But interclade snps are already known for much of that sector; donor selection will be important to guide the search toward finding the really needed new snps in that region.

I'm I1-ASgeneric; but I'm encouraged enough to contribute something to any of these four sector searches.

******* All that is needed are organizers to step forth to put the syndicates together and get the orders into the lab. ******

I am encouraged that Krahn intends to run all donors through the same 100,000 nucleotide sites for some indefinite period. This means an entire tree sector campaign does not need to be decided out at the very beginning. Future donors from various sectors can be tested at later times with the information learned from the earlier donor test results and all can be compared against each other. This is more powerful searching if carried out in a reasoned way. The initial investigations in each sector can go forth with just 2 well-chosen donors.


p.s. All estimates above are based on an assumed nucleotide site mutation rate of 2 / (100 million) If the lab now has a better number for this, chances of success scale up or down accordingly. I don't know what they know about that number.

This thread: