GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2007-12 > 1198951540


From:
Subject: Re: [DNA] Any news on the "Walk Through the Y"initiative??(Theprocess, & a proposal)
Date: Sat, 29 Dec 2007 10:05:40 -0800


> So in order to find one
> desired SNP with some confidence a million nucleotide sites need to be
> scanned and compared between the four dna samples. Is this financially
> plausible today?
>
> Ken

In response to your question, I'll quote Richard Kenyon's post from
yesterday (Where's the SNPs?) -

> A lot can be learned from Dr. Thomas Krahn's website at
> ymap.ftdna.com . Put ChrY in the search box to see a map of the
> entire 57,772,955 nucleotides of the Y-chromosome. Checking the box
> labeled dbSNP displays a graph of the number of known SNPs per 20
> Kb along the length of the Y-chromosome. There are other boxes that
> can be checked, such as CV's private SNPs (Craig Venter's) and JW's
> private SNPs (James Watson's).
> Surprisingly, most of the SNPs we use are clustered around y-
> positions 13-14 and 20-21 megabases, measured from the tip of the p-
> arm of the chromosome. As can be seen from the dbSNP graph, both of
> these regions have very low SNP density (i.e., number of SNPs per
> 20 kilobases). The highest SNP density is between postions 3 and 7
> kb. I don't know if this is significant or not. It does seem that
> each research group had most of the SNPs that they reported were
> clustered.

Your concern is one of these reasons that I suggest that we start
with the most promising clusters, those that demonstrate very, very
clear modals, & that we carefully choose the most qualified
participants from within the those clusters. To put it bluntly, we
could get lucky, & that's what I'm hoping for. It's remotely possible
that we could choose the most likely segment of DNA in two of the
most promising candidates from the most promising clusters, (drum
roll), & find the hoped for results in the first round. The odds are
certainly better than the California Lottery. Even so, yes, there is
certainly a risk of a long term commitment to a slow process. The
very worst case scenario is that you run through the entire Y & find
your results on the last segment tested. Actually, the very worst
case scenario is that technical problems lead to catastrophic failure
of an entire segment sequence, & those candidates go to the back of
the line.

By the way, Ken, if you are really worried about the taxi scenari,
bring a Garmin nuvi 680 GPS with you, & if your taxi driver is
straying from the straight & narrow, get him back on the right path.
Use the detour notification service (annual subscription) to avoid
traffic jams. You can get one at Costco for $479, plus tax (first
year's subscription included). Of course that's a lot more than
you're likely to pay even the most predatory taxi driver, but what
you are paying for is peace of mind, *& the conviction that you know
where you're going & how to get there*. Technology is expensive, but
that's what I'm willing to pay for, a commitment to knowing where I'm
going. (Actually, I have the Garmin nuvi 200, & it's suits my
purposes just fine. Almost all of the taxi drivers use GPS now, so
it's hard for them to pretend they're lost.)

By way of a more direct response, the cost of SNP discovery has been
dropping dramatically over the past two years, & I think that it is
within range. Pat

This thread: