GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2006-08 > 1155350997


From: "William Hurst" <>
Subject: New mtDNA Fluxus diagram from K Project
Date: Fri, 11 Aug 2006 22:49:57 -0400


Hi all,

I have created another Fluxus diagram from the mtDNA Haplogroup K Project
results. This one includes 36 sequences (of the current 215) which have HVR2
mutations 146C and 152C in addition to the basic six K differences from the
CRS. So the basic sequence is 16224C, 16311C, 16519C, 73G, 146C, 152C, 263G
and 315.1C. The diagram is published at
http://freepages.genealogy.rootsweb.com/~wrhurst/mtdna-k/eitheror.jpg For a
reason to be explained, I have named this sequence "either/or."

Starting with the Kroot node at the lower left, follow the 146C and 152C
mutations up to the basic either/or sequence at the largest node, which is
labeled 9893 after the first of six matching haplotypes. One person at that
node has received the subclade designation K2a, based on a full-sequence
test. You might suspect that the other five with perfectly matching HVR
haplotypes would also be K2a's - but not necessarily. Next look to the right
past mutation 512C to a slightly smaller node labeled N17958 representing
four persons who are probably in subclade K2a2a, one of the Ashkenazi
subclades defined in Dr. Doron Behar's paper at:
http://www.familytreedna.com/pdf/43026_Doron.pdf Logically, K2a2a follows
K2a. Note that there are several additional coding-region mutations which
define K2a2a; but those will have to be assumed, since none of the four has
taken the full-sequence test. The assumption is a good bet as 512C does not
appear elsewhere on Behar's K charts.

However, moving up from the large node past mutation 498- (a deletion at
HVR2 position 498) leads to a node labeled N18286. The 498- mutation, along
with 146C and 152C, is shown on Behar's chart as defining K1c. There is
certainly no logic for K1c to be downstream from K2a! In fact, on Behar's
chart, the 146C, 152C, 498- sequence and the 146C, 152C, 512C sequence are
completely separate. In a way, Fluxus confuses the issue; but it also
demonstrates a fact: identical HVR sequences are not necessarily in the same
subclade. The subclades of the other five persons in the 9893 node are
unknown and would only become known by full-sequence tests. I suspect some
of them would be assigned to subclade K1*, although "pre-K1c" might be more
accurate. So they are either K1* or K2a. Some of them, and 10 other
haplotypes with additional mutations branching off the largest node, may be
in subclades downstream from K2a other than K2a2a.

Branching off N18286, the K1c node, past the 16320T mutation, is found the
four-member perfect K1c2 node marked 58959, from which radiates three other
K1c2's with additional mutations. There are also seven other haplotypes
branching directly off the K1c node, some of which may be in K1c1 or its
lower subclades, based on coding-region mutations.

One should note that in all of these haplotypes there is not a single
insertion at HVR2 position 524. I have previously said that the 524's are
"recurrent but not random." I have never seen a 524 insertion in any of the
Ashkenazi subclades or in the K1c subclades.

Someday in the future we should have better answers when either a subclade
test or a cheaper full-sequence test is available. Within the 36 persons on
this chart there are only two with subclade assignments based on
full-sequence tests.

For other haplogroup K Fluxus diagrams and other information, see the News
tab on the mtDNA Haplogroup K Project website at
http://www.familytreedna.com/public/mtDNA_K/ The K Project may be joined by
those in mtDNA haplogroup K by clicking on the Join button on your FTDNA
personal page.

Bill Hurst
Administrator, mtDNA Haplogroup K Project


This thread: