GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2007-12 > 1197886255


From: "Tim Janzen" <>
Subject: Re: [DNA] SNPs as Clock
Date: Mon, 17 Dec 2007 02:11:00 -0800
In-Reply-To: <BAY103-DAV7E9B8C81FCC01E3B57309C8620@phx.gbl>


Dear Richard,
I agree that as we start doing testing of large blocks of the Y
chromosome for multiple people that we will need to use caution when
interpreting results from near the centromere or from the tips of the p and
q arms since those areas apparently recombine with each generation. Like
you, I'm not sure of the exact limits of the non-recombining portions of the
Y-chromosome. Gareth Henson's Excel file which shows Ventor's SNPs at
http://freepages.genealogy.rootsweb.com/~ncscotts shows three MNPs in the
first 2500 base pairs of the p arm and then there are no SNPs until the 2.72
million base pair position. I don't know if they didn't sequence the p arm
between the 2500 base pair position and the 2.72 million base pair position
or what happened in that portion of the sequencing. In any case, the bulk
of the SNPs in Ventor's Y chromosome appear be in the p arm. Note that they
didn't sequence Watson's Y chromosome between 1 and 2.72 million base pair
position either (or at least there were no SNPs found there). I suspect
that the first 2.72 million base pairs recombine with each generation and
thus they won't be helpful regions in which to look for SNPs.
Here is a breakdown of the SNPs, MNPs, mixed sequence variants,
insertions, and deletions in Ventor's Y chromosome broken down by base pair
position:
1-2429: 3
2429-2.72 million: uncertain
2.72-3 million: 101
3-4 million: 1787
4-5 million: 2152
5-6 million: 1578
6-7 million: 968
7-8 million: 21
8-9 million: 19
9-10 million: 14
10-11 million: 209
11-12 million: 159
12-13 million: 213
13-14 million: 75
14-15 million: 21
15-16 million: 23
16-17 million: 35
17-18 million: 26
18-19 million: 56
19-20 million: 53
20-21 million: 604
21-22 million: 32
22-23 million: 95
23-24 million: 127
24-25 million: 87
25-26 million: 11
26-27 million: 37
27-28 million: 31
28-57.39 million: 0 (apparently not sequenced since it is
heterochromatin)
57.39-57.44 million: 684

In comparing the above analysis with the analysis I did for Watson's
DNA I note the following:
1. Watson seems to be more closely related to the HUGO reference standard
for the Y chromosome at
http://www.ncbi.nlm.nih.gov/mapview/maps.cgi?taxid=9606&chr=Y than Ventor
is. This is evidenced by the fact that Ventor has over 4x more SNPs as
compared to the HUGO reference standard than Watson does (8634 versus 1746).
2. If you compare the summary above with the summary for Watson's SNPs I
wrote at
http://archiver.rootsweb.com/th/read/GENEALOGY-DNA/2007-09/1189709479-01 you
will note that these three portions of the Y chromosome seem to consistently
have more SNPs than other portions of the Y chromosome:
3-7 million base pair region
10-14 million base pair region
57-58 million base pair region
Thus it would likely be productive to sequence large portions of the
3-7 million base pair region and the 10-14 million base pair region when
looking for deep subclade SNPs and family SNPs. It would probably be best
to avoid the 57-58 million base pair region when looking for these kinds of
SNPs since that area may combine with each generation.
3. The ratio of SNPs per million base pairs is not consistent for both
Watson and Venter relative to the HUGO reference standard. Watson
proportionately has more SNPs in the 10-14 million base pair region than
Ventor does. However, Ventor proportionately has more SNPs in the 2.7-7
million base pair region than Watson does.
4. There is an interesting spike in the number of SNPs between the 20-21
million base pair region in Ventor's DNA that isn't seen nearly as clearly
in Watson's DNA. Perhaps Peter Underhill and his colleagues knew something
about this region that we didn't before he undertook his study of it. His
primary paper is found at
http://www.human-evol.cam.ac.uk/Members/Lahr/pubs/AHG-65-01.pdf.

Sincerely,
Tim Janzen

-----Original Message-----
From:
[mailto:] On Behalf Of RICHARD KENYON
Sent: Monday, December 17, 2007 12:01 AM
To:
Subject: Re: [DNA] SNPs as Clock

Tim,
Thanks for the results of your analysis of SNP loci. Some months ago I
extracted SNP details from Thomas Krahn's ymap website, placing the
information in an MS Access database. Your analysis certainly shows quite
clearly that Underhill's group concentrated their analysis almost entirely
to the 20-21 million region.
The SNPs in non-recombining portions of the Y-chromosome are the ones
that of interest. I'm not sure of the exact limits of the non-recombining
portion of the Y-chromosome, but I know that regions around the centromere
and telomeres are subject to recombination so they should be avoided.
Incidentally, ymap can display the SNPs for Watson (jw) and Venter (cv).
I assume that most of these are private SNPs.

Richard R. Kenyon ("Dick")

This thread: