GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2007-01 > 1167685561


From: (John Chandler)
Subject: Re: [DNA] DYSDYS388 mutation rate [was Re: Freq's 459Unique R1b Iberian Haplotypes]
Date: Mon, 1 Jan 2007 16:06:01 -0500 (EST)
References: <bd2.c5ee536.32c90b09@aol.com> <4597D720.7040707@sbcglobal.net><000801c72cfe$3bdbe2d0$6400a8c0@Ken1> <REME20061231152748@alum.mit.edu>
In-Reply-To: <REME20061231152748@alum.mit.edu> (john.chandler@alum.mit.edu)


All,

Linkage disequilibrium is a major effect on Y chromosome haplotypes.
I made the following table based on a comparison of the modes and rates
determined from two disjoint subsets of my Y37 dataset. Subset 1 consists
of all haplotypes with DYS388<=12, while Subset 2 has all haplotypes with
DYS388>12. The rate estimates for most markers differed sharply between
the two subsets, not just for DYS388. The columns marked "Md1" and "Md2"
show the modes for the two subsets; "Dif" shows "+" if Subset 2 has a
significantly higher rate estimate than Subset 1; "-" if a significantly
lower rate; and "=" if approximately the same rate. The column marked
"Corr" shows which markers have a positive correlation between mode and
rate, i.e., those where the mode is higher for the subset with the higher
rate estimate. Of the 30 loci included, 19 have positive correlation, 9
have either the same mode or roughly equal rate estimates, and only 2 have
negative correlations. The comparison for DYS388, which should have the
greatest contrast, is indeed striking: in units of 10^-5 per generation,
Subset 1 has a rate of 5 +/- 1, while Subset 2 has 105 +/- 17.

[The table is formatted to be aligned correctly when viewed in a fixed-
width font, but should be reasonably well-aligned in any font.]

Locus _ Md1 _ Md2 _ Dif _ Corr
393 ___ 13 __ 13 ___ + ___ ?
390 ___ 24 __ 23 ___ - ___ y
19 ____ 14 __ 14 ___ + ___ ?
391 ___ 11 __ 10 ___ - ___ y
385 ___ 11 __ 14 ___ + ___ y
426 ___ 12 __ 11 ___ = ___ ?
388 ___ 12 __ 14 ___ + ___ y
439 ___ 12 __ 11 ___ - ___ y
389i __ 13 __ 12 ___ - ___ y
392 ___ 13 __ 11 ___ - ___ y
389ii _ 16 __ 16 ___ = ___ ?
458 ___ 17 __ 15 ___ - ___ y
459 ___ 09 __ 09 ___ - ___ ?
455 ___ 11 __ 11 ___ = ___ ?
454 ___ 11 __ 11 ___ = ___ ?
447 ___ 25 __ 23 ___ - ___ y
437 ___ 15 __ 16 ___ + ___ y
448 ___ 19 __ 20 ___ - ___ n
449 ___ 29 __ 28 ___ - ___ y
464 ___ 15 __ 14 ___ - ___ y
460 ___ 11 __ 10 ___ - ___ y
H4 ____ 11 __ 10 ___ - ___ y
YCAII _ 19 __ 19 ___ - ___ ?
456 ___ 16 __ 14 ___ - ___ y
607 ___ 15 __ 14 ___ - ___ y
576 ___ 18 __ 16 ___ - ___ y
570 ___ 17 __ 18 ___ + ___ y
CDY ___ 37 __ 35 ___ - ___ y
442 ___ 12 __ 12 ___ - ___ ?
438 ___ 12 __ 10 ___ + ___ n

Another table that might be of some interest: the histograms for the
two subsets:

388 ( 4-12)= 1 0 0 1 1 1 61 29 6190
388 (13-18)= 761 925 287 144 28 1

Note: as I mentioned earlier, I also looked at a subset with DYS388>=12.
The rate estimate for DYS388 was 38 +/- 6 (10^-5), which is close to the
rate estimate based on the combination of my Y37 and Y25-not-37 datasets.

Further note: the rate estimates for all subsets were computed using the
same father-son rate calibration data, which included some for DYS388.
This should have the effect of boosting the estimates for all markers with
Subset 2 if all else were equal. Nonetheless, many more markers showed a
smaller rate with Subset 2 than showed a larger rate.

John Chandler

This thread: