AS OF APRIL 13, 2006


This is my fifth survey of the mtDNA haplogroup K (Katrine’s Clan) entries on FamilyTreeDNA’s MitoSearch. The previous survey at 403 entries may be found at K403 Survey. There are several links on that page to the previous survey, to a chart of the high-resolution entries, and to a supplement on geographical considerations of the entries. I will not repeat everything discussed in those documents.


The K500 survey includes a new CHART, which is sorted by HVR2 then HVR1 mutations. It took five months for the entries to go from 100 to 200, but only five weeks to go from 400 to 500. At some point, I will start skipping the century marks for these surveys.


Only non-duplicated high-resolution or HVR1 plus HVR2 entries are included in this survey, in which are found 230 entries – or 46% of the total, only slightly higher than in the previous survey. Among those removed were 22 claiming to be CRS, or having no HVR1 mutations. That is highly unlikely for haplogroup K. None of the 230 included entries has fewer than three HVR1 mutations. I also removed one duplicate entry. Twenty-five of the remaining entries are marked Show, which means they have a pedigree chart on MitoSearch. The number of those testing outside FTDNA has not increased; all but five tested with FTDNA. There are still five fairly-obvious K’s under Unknown category at MitoSearch who tested with Relative Genetics. There are also five U5 entries and one U4 entry which may eventually be reclassified as K’s; some of those have perfect matches with K entries.


The 230 entries include 154 different haplotypes, for a 67% diversity. That’s a decrease from last time’s 78%. That means that new entries are becoming more likely to find matches. There are 123 “singleton” entries. Three haplotypes have eight examples, and two each have six and seven. For some reason, there are single sets of four and five examples. Four have three, and there 18 paired haplotypes. The multiple-entry haplotypes are easily found on the accompanying chart.


I have again assigned country codes to the entries based on the listed country of origin. New countries this time are Bahamas and Moldova. Outside the USA, the highest area representation is still by the UK with 26, but that’s divided between England, Scotland, and Northern Ireland. Ireland itself has 18 listings. There are 34 who listed one of the Eastern European countries. The only other country with double-digits entries was Germany with 11. In the USA, the highest number came from southeastern states with 33. In the geographic supplement to the previous survey, I suggested a correlation between those states and the British Isles. A full 49 entries listed the county of origin as Unknown.


I have used the same colors to mark certain mutations as in the previous charts. Yellow denotes mutations 498- or 16320T, which probably predict subclades K1c and K1c2. Green marks the various “Ashkenazi” mutations, 16223T, 16234T, 16524G, and 512C, as discussed in Dr. Doron Behar’s paper. Blue marks 497T, which starts all K1a and lower subclades. Orange marks a few entries with 16270T, which is the defining mutation for haplogroup U5; those are the ones which have U5 matches. Aqua marks entries with the 133G mutation, which is found in most of the “odd” haplotypes with several probable back mutations. There are no new examples of this group, which was discussed in the previous geographic supplement.


In the previous geographical supplement, I had presented a few apparent discoveries. The various “Ashkenazi” mutations occurred mostly in those listing one of the Eastern Europe countries as their origin. Those with the 16223T and 16234T mutations generally came from all the Eastern Europe countries. But those with the 512C mutation came only from the northern Eastern Europe countries Russia and Lithuania. Since then one entry from Belarus has been added. The 16524G mutation came from southern Eastern Europe countries including Poland, Romania, Ukraine and Hungary. Now Moldova has been added. There is actually another new entry with 16524G from Norway. However, since this entry has four of the HVR2 524 insertions, which otherwise have not occurred with any of the “Ashkenazi” mutations, I suspect this 16524G mutation may be of the “personal,” “family” or “random” category. If the theory holds up in the future, a person with either the 16524G or 512C mutations who can only list a USA state or Unknown for their origin will have at least some idea of what part of Eastern Europe their ancestor came from.


I had noticed that the 498- mutation without the 16320T mutation – generally subclade K1c – was found only in those with origins from the island of Great Britain, while 498- with 16320T – K1c2 – was found in Ireland, Scandinavia and Western Europe. Although this still holds true, the sample is still too small to try to prove some migration pattern.


The new feature on the accompanying chart with this survey is three new charts at the bottom of the main chart. The first one is a bar chart with the actual counts of the listed countries of origin for the 230 entries. To me, the main point visualized is the great number of K’s who are brickwalled in the southeastern United States. The first pie chart shows the origin percentages by major areas. Here almost exactly half are brickwalled in North America or Unknown. The second pie chart shows just the European areas of origin. While British Isles is the largest area, I think it would be even larger if those brickwalled folks mentioned above knew their country of origin. Immigration from other areas, such as Eastern Europe, was probably recent enough so that most know their country of origin.


All K’s who tested at FTDNA or who transfer their results from the National Geographic Society Genographic Project are welcome to join the mtDNA Haplogroup K Project. Further information is on our website. At 127 members, we remain the largest mtDNA haplogroup project.


Three caveats: One, usage of the term Ashkenazi to describe certain mutations or haplotypes does not mean that everyone with those mutations has Ashkenazi ancestors. Two, mention of subclades is for descriptive purposes only. Subclade definitions usually require a subclade test of the coding-region mutations outside the hypervariable regions (HVR) or a full-sequence test. As of now, FTDNA does not offer a subclade test for K. A few K’s have taken the expensive full-sequence test, but those results are not reflected in the data extracted from MitoSearch. Three, MitoSearch entries are of course USA-centric and are not necessarily representative of the world’s K population. However, MitoSearch probably has a better representation of British-origin K examples than other available studies.


William R. Hurst