GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2007-03 > 1172801837


From: James Heald <>
Subject: Re: [DNA] Challenge to YDNA J Gurus
Date: Fri, 02 Mar 2007 02:17:17 +0000
References: <001901c75c2a$8ba69340$6400a8c0@Ken1>
In-Reply-To: <001901c75c2a$8ba69340$6400a8c0@Ken1>


Here's a distance matrix, and central haplotypes, for J1. I hope that
people's mail readers don't wrap lines at 80 columns!

Haplotypes are based on those defined by Bonnie Schrack for the Y-DNA-J
project. Dataset is that project + haplogroup J1 hits at Ysearch, with
duplicates (and in a couple of cases close relatives) excluded.

Distances = (1/37*100) * Sum{ (Mismatch^2)/ Mutation Rate },
so the numbers should roughly represent hundreds of generations (but
likely to be rather noisy, and -possibly- systematically *low*; the
latter needs more research. I have no idea what size (or skewness) of
error bars are appropriate).

Numbers are for (squared) distances, ie /not/ halved to give symmetric
times.

The "var" gives the average squared distances inside the clusters from
the cluster centres -- ie a rough and ready estimate of the TMRCA of the
cluster itself (assuming a "star" tree shape, which is probably deeply
unrealistic).

Mutation rates are Chandler (2006) with the exception of DYS 388 set to
0.00225, rather than 0.00022, reflecting that it seems to mutate much
more often at the high repeat numbers found in Haplogroup J.

On a principal components plot, cluster J1-STR2 notably stands out apart
from all the rest.

The other clusters used to be separated really well, if for empirical
relative mutation rates I set the rates of all of FTDNA's "slow" markers
to 1, and al of the "fast" markers to 4.

But with the Chandler numbers, STR clusters 7,8,9 and 9a are all very
close. In particular the centres of 7 & 8, 8 & 9 and 3 & 9 are closer
to each other than they are to average members of their clusters.
Tomorrow I'll do a principal components plot just for these groups, and
see if there is any other clustering that jumps out from the plot,
assuming that we believe in the Chandler rates.

Only the markers up to DYS 438 were used in the calculations; after that
beware that the central haplotypes may be estimated on the basis of, in
some cases, only one or two samples.

Finally, for those who are interested in such things, clusters 7,8,9 and
9a match the infamous six marker Cohen Modal Haplotype. The small group
of probable Ashkenazi Cohens on public databases I've seen are in
cluster 9a; but these may well share a common ancestor in Medieval times.

All best,

James.


J1-S1 J1-S2 J1-S3 J1-S4 J1-S5 J1-S6 J1-S7 J1-S8 J1-S9 J1-S9a
J1-S1 0 11.34 9.25 6.58 6.9 7.02 7.6 9.05 8.77 10.48
J1-S2 11.34 0 7.55 6.64 5.48 7.78 5.58 5.65 6.98 7.79
J1-S3 9.25 7.55 0 3.78 2.87 4.82 2.44 1.48 0.63 1.91
J1-S4 6.58 6.64 3.78 0 1.75 2.47 2.75 3.54 3.67 5.2
J1-S5 6.9 5.48 2.87 1.75 0 4.74 1.57 2 2.66 3.97
J1-S6 7.02 7.78 4.82 2.47 4.74 0 3.98 5.4 4.47 5.74
J1-S7 7.6 5.58 2.44 2.75 1.57 3.98 0 1.13 1.52 1.42
J1-S8 9.05 5.65 1.48 3.54 2 5.4 1.13 0 0.89 1.28
J1-S9 8.77 6.98 0.63 3.67 2.66 4.47 1.52 0.89 0 1.03
J1-S9a 10.48 7.79 1.91 5.2 3.97 5.74 1.42 1.28 1.03 0

Var 1.22 0.26 1 0.12 0.96 1.64 1.52 1.51 1.31 0.47




393 12 12 12 12 12 12 12 12 12 12
390 24 25 23 23 23 23 23 24 23 23
19 13 14 14 14 14 14 14 14 14 14
19b
391 10 10 11 10 10 10 10 10 10 10
385a 12 16 13 13 13 13 13 14 13 13
385b 19 17 18 19 18 19 16 16 17 15
426 10 11 11 11 11 11 11 11 11 11
388 15 16 17 15 17 13 16 17 16 16
439 13 10 11 13 12 12 12 12 11 12
389|1 12 13 13 14 13 13 13 13 13 13
392 11 13 11 11 11 11 11 11 11 11
389|2 17 16 17 17 18 16 17 17 17 17
458 18 17 19 17 17 20 18 18 18 17
459a 8 8 8 8 8 8 8 8 8 8
459b 9 9 9 9 9 10 9 9 9 9
455 11 11 11 11 11 11 11 11 11 11
454 11 11 11 11 11 11 11 11 11 11
447 24 26 26 26 25 27 25 25 25 26
437 14 14 14 14 14 14 14 14 14 14
448 20 20 20 20 20 21 21 20 20 21
449 32 28 26 26 26 29 27 25 26 26
464a 14 13 12 13 13 15 12 13 12 12
464b 15 14 14 15 13 15 14 14 14 14
464c 16 14 16 16 15 16 15 16 16 16
464d 16 17 17 16 17 16 16 16 17 17
464e
464f
464g
460 10 11 10 11 11 11 12 11 11 11
H4 9 9 10 10 10 10 10 10 10 10
YCAIIa 19 19 22 19 19 20 20 21 22 22
YCAIIb 21 22 22 22 22 22 22 22 22 22
456 15 17 14 14 16 15 15 15 15 15
607 13 14 14 12 14 13 14 14 14 14
576 17 18 18 16 17 17 19 18 18 19
570 17 18 18 18 18 17 18 17 17 18
CDYa 32 32 32 33 33 34 32 32 33 32
CDYb 38 37 35 36 35 36 36 34 36 35
442 14 12 12 12 12 12 12 12 12 13
438 10 11 10 10 10 10 10 10 10 10
425 11 12 12 12 12 12 12 12
461 11
462 12
A10 12
635 22
1B07 11
441 14
444 12 12 12 12 13 12 12 12
445 11
446 14 14 15 14 15 14 14 13
452 10
463 19
531 11 11 11 11 12 13 11 11
578 8 8 8 8 8 8 8 8
395S1a 15 15 15 15 15 15 15 15
395S1b 16 16 16 16 16 16 16 16
590 8 8 8 8 8 8 8 8
537 12 11 11 11 11 12 11 11
641 10 10 10 10 10 10 10 10
472 8 8 8 8 8 8 8 8
406S1 12 11 11 11 11 13 11 11
511 9 9 9 9 9 9 9 9
413a 22 21 21 21 21 21 21 21
413b 22 23 22 22 22 21 22 22
557 18 16 19 19 21 19 19 18
594 10 10 10 10 10 10 10 10
436 12 12 12 12 11 12 12 12
490 12 12 12 12 13 12 12 12
534 17 16 15 16 17 16 15 15
450 8 8 8 8 8 8 8 8
481 26 26 23 25 25 25 25 24
520 23 20 21 21 20 20 21 21
617 12 12 12 12 12 12 12 12
568 12 11 11 11 12 11 11 12
487 14 14 14 15 14 13 13 14
572 11 12 12 12 11 11 11 12
640 12 11 12 12 12 12 12 12
492 12 12 12 12 12 12 12 12
565 11 11 11 11 11 11 11 11
434 9
435 11
485 15
494 10
495 15
505 12
522 11
533 11
549 12
556 12
575 10
636 1 3
638 10
643 10
714 25
716 26
717 19
726 12
156-Y


This thread: