GENEALOGY-DNA-L Archives

Archiver > GENEALOGY-DNA > 2005-09 > 1126977672


From: "Ken Nordtvedt" <>
Subject: Re: [DNA] mtDNA Haplogroup K in mitosearch - Update
Date: Sat, 17 Sep 2005 11:21:12 -0600
References: <BAY104-DAV17923D032E6D025ECAC8D6F1310@phx.gbl> <BAY104-DAV612682E9314C72FFBD0CBF1900@phx.gbl>


Any geographical patterns seen? All I know is about the "ice man" found in
the Alps and a few thousand years old.
----- Original Message -----
From: "William Hurst" <>
To: <>
Sent: Saturday, September 17, 2005 10:51 AM
Subject: [DNA] mtDNA Haplogroup K in mitosearch - Update


> Hi all,
>
> Back on April 10 I gave a little report on the first 100 mtDNA haplogroup
> K (Katrine's Clan - spelling is important) entries on mitosearch. I have
> attached the original report below so I don't have to explain everything
> again. If you are really interested, you might want to read that first.
> Some of my statistics are better, or at least different, since I have
> learned how to export the data to Excel and manipulate it there.
>
> I noticed yesterday that there are now 201 K entries, doubled in five
> months. K is a good haplogroup to study since it is not too big with lots
> of subclades, like H, or so small there are few entries. It's just right.
> Of course, it's my Clan.
>
> Of the 201, most tested with FTDNA of course, but a few tested elsewhere.
> Four from the NGS Genographic Project. Three from Oxford Ancestors. One
> from Sorenson - hmm, how did the person get the data? One from DNA
> Heritage - hmm again, I didn't think they even offered mtDNA tests. Three
> from "Other." (Relative Genetics offers mtDNA tests, but none were in K;
> apparently they don't tell you your haplogroup, so you have to look under
> the "Unknown" category to find them. By the way, there are now four
> K-looking entries under Unknown.)
>
> There are now three CRS interlopers in K. I have not looked at them
> further. I also eliminated those who do not have data for HVR2. I consider
> "No Mutations" the same as "Not Tested," since every haplotype with HVR2
> data has at least three mutations. So I ended up with 104 entries with
> HVR1 plus HVR2.
>
> Of the 104, there are 83 different haplotypes. That's a "diversity
> percentage" (I think it's called) of 79.8. 71 of these are singletons and
> only 12 with multiple examples. One haplotype has six entries; one has
> five. All the others only have two or three.
>
> The motifs for K are 16224C and 16311C in HVR1. Two of the entries are
> missing 16311C, which may be due to back mutations or just mistakes by the
> persons entering the data.
>
> All 104 also have 16519C. That mutation may be called a "hotspot," but
> that just means it shows up in many haplogroups. It obviously doesn't mean
> it is a fast-mutating marker. (There are some HVR1-only entries which
> don't have 16519C, but those are usually from, say, Oxford Ancestors,
> which doesn't test that high in HVR1.)
>
> All but two have the HVR2 mutation 073G. Also very common are HVR2
> mutations 263G and 315.1C.
>
> As I said, there are no generally recognized subclades of K. Several
> researchers have used different definitions for a K1; K2 is more or less
> standardized. (These definitions don't match those on list-member Ian
> Logan's website, which uses full sequences which apparently don't include
> anybody from or near Scotland.)
>
> The most liberal definition of K1 includes the HVR2 mutations 146C and
> 152C. There are 15 examples of that. John Walden's definition (see URL in
> previous message below) requires the deletion 498-; there are six of
> those. If 16320T is required, there are nine more. There is one person who
> has 16320T and 146C, but not 152C and 498-. So there are 31 who might be
> called K1. The haplotype with the most examples, six, is here and includes
> in full 16224C, 16311C, 16320T, 16519C, 073G, 146C, 152C, 263G, 315.1C,
> 498-.
>
> K2 under most definitions has the motif 16093C. There are 19 of those.
> (One person, me, has all the motifs for K1 and K2; John Walden has told me
> I'm a K1 with the 16093C as just a random addition.)
>
> There are two different combinations which, according to list member Ellen
> Coffman, are restriced to Ashkenazi Jews; 16234T with three examples, and
> 16234T and 16223T also with three examples.
>
> Looking at John Walden's K chart, it seems that another subclade should be
> defined by 497T. There are 57 entries with that mutation. None of the K1
> subclade persons have 497T. Eleven have 497T and 16093C, the motif for K2.
> I'll let somebody else figure out if these are K2's with an additonal 497T
> or the unnamed 497T-defined subclade with an additional 16093C. The second
> most common haplotype, with five examples, is found here. In full: 16224C,
> 16311C, 16519C, 073G, 263G, 309.1C, 315.1C, 497T.
>
> Bill Hurst
>
>> [April 10, 2005 report]
>> I've been waiting for this day. The number of K - Katrine Clan - entries
>> in FamilyTreeDNA's mitosearch has just reached 100. (OK, there are
>> actually 101, but one is a stray CRS. I chased away another CRS weeks
>> ago, but this one is still there. Also, there are at least two K-looking
>> entries in "Unknown" which I haven't counted.) K people comprise only
>> about 8% of the European-derived population, but we seem to take up more
>> of the List discussion than that.
>>
>> Of the 100, 62 have entries for both HVR1 and HVR2. Of the 38 HVR1-only
>> entries, 35 have HVR2 marked "Not Tested" and 3 are marked "No
>> Mutations." I seriously doubt if there are any K's with no HVR2
>> mutations, so I've treated both the same.
>>
>> The motifs for K are 16224C and 16311C. All 100 have the former; 1 entry
>> is missing 16311C, which might be a reverse mutation or a typographical
>> error. (Remember that in some cases the mutations may have been entered
>> by hand, although most were probably entered by one click from the
>> personal page.)
>>
>> Of the 100, all but 4 have 16519C. Of those, 3 were tested by Oxford
>> Ancestors which does not test for that mutation. So only 1 of 97 is
>> really missing 16519C, which is why I have said it is not "unstable," at
>> least for K.
>>
>> There are several mutations common in HVR2, usually as a result of the
>> CRS sequence not being the most common at those locations. 60 out of 62
>> have 073G. Three of 62 are missing 263G, but one of those has a unique
>> 253G. Typo perhaps? One is missing both 073G and 263G. All but one of 62
>> has the insertion 315.1C.
>>
>> The shortest complete sequence found was 6 mutations - 3 each in HVR1 and
>> HVR2. Only that entry listed as few as 3 HVR2 mutations. One sequence has
>> only 2 in HVR1 - the one missing 16311C.
>>
>> Two share the honor of the most mutations - 16. One has 7 HVR1 mutations;
>> the other has 12 HVR2 mutations - both records for those regions.
>>
>> Of the full 100, 32 have the HVR1 sequence 16224C, 16311C, and 16519C.
>> Adding HVR2 to those results in only 2 pairs of exact matches.
>>
>> I did not do an extensive search for exact matches, but I think the
>> record for high-resolution perfect matches is 3 entries with 16224C,
>> 16311C, 16320T, 16519C, 073G, 146C, 152C, 263G, 315.1C, and 498-. A
>> perfect generic K1 subclade sequence, as discussed below.
>>
>> K is often given two subclades: K1 and K2. There is no official
>> definition for these. K2 is usually identified by having 16093C. There
>> are 21 entries with that mutation. One is not a K2, as I will discuss
>> later. So 32% of the complete sequences are K2.
>>
>> I have seen three different motif lists for K1 which fit the mitosearch
>> entries. (The chart prepared by Ian Logan has K1 and K2, but the
>> sequences used are very different from those in mitosearch and are
>> defined mainly by coding region mutations. John S. Walden's K chart at
>> http://freepages.genealogy.rootsweb.com/~jswdna/kchart.gif<http://freepages.genealogy.rootsweb.com/~jswdna/kchart.gif>;
>> on the left side shows the connections between the motif mutations for
>> K1.)
>>
>> The most liberal definition of K1 requires 146C and 152C in HVR2. 19 or
>> 32% have both. 28 have just 146C and 21 have just 152C.
>>
>> John Walden has suggested the motif for K1 should be the deletion 498-.
>> 11 or 17% of the entries have that one. All 11 also have 146C and 152C.
>>
>> The most common definition of K1 is the 16320T mutation. Nine of the 100
>> have that mutation.
>>
>> Seven entries have 16320T, 146C, 152C and 498-. Only 1 with 16320T, who
>> tested for HVR2, didn't have either of the other motifs. Perhaps that was
>> a parallel mutation. K1 is often called "Scottish." Of the 4 entries on
>> mitosearch with the Origin given as Scotland, all have at least one of
>> the motifs for K1.
>>
>> Recently discussed on the List were Ashkenazi Jewish K sequences, given
>> as 16223T, 16234T, or both. Only 5 have 16223T; only 7 have 16234T; 4 of
>> those have both. None of those has the motifs for K1 or K2. Five list
>> their Origin as Russia or Poland.
>>
>> Several other mutations are found in K in reasonably large numbers, but
>> don't seem to point to any specific subclade. In HVR2, 34 have 497T, 20
>> have 309.1C, 17 have 195C. Looking at Walden's chart and the number here,
>> it seems as if 497T should define a separate subclade.
>>
>> There are 4 different combinations of insertions of 524 in HVR2. Two
>> entries have 524.1C and 524.2A. Seven others add to that 524.3C and
>> 524.4A. Four have 524.1A and 524.2C. Two more have those and add 524.3A
>> and 524.4C. All but one of those 15 also have 497T.
>>
>> As I said earlier there is one entry with 16093C, usually the motif for
>> K2, but with all the motifs for K1. That entry also has a unique - and
>> apparently not just for K - 081.1T mutation. That one is mine, of course,
>> perhaps why I find this subject so fascinating.
>>
>> Bill Hurst
>
>
> ==============================
> View and search Historical Newspapers. Read about your ancestors, find
> marriage announcements and more. Learn more:
> http://www.ancestry.com/s13969/rd.ashx
>
>


This thread: