Loading ...
Sorry, an error occurred while loading the content.

Re: Current Research R1a and Jewish R1a subclades

Expand Messages
  • aj_levin
    Hi Q cousins, and very sorry for your loss, Rebekah. This is a slight edit of my earlier post just to add it to the original thread as a reply. R1a is still
    Message 1 of 2 , Aug 24, 2010
    • 0 Attachment
      Hi Q cousins, and very sorry for your loss, Rebekah. This is a slight edit of my earlier post just to add it to the original thread as a reply.

      R1a is still not too well understood, or much studied compared to R1b, but we've been making some progress. A good background read of some of the STR and SNP groupings is to be had here:


      The following Family Tree DNA projects are useful for research:



      Wikipedia should usually be read skeptically, but this page is helpful:


      The draft tree here is also invaluable, though a little hard to read because of a mix of ISOGG and YCC terminology, plus discoveries not yet incorporated into either system which mean multiple levels still have the same name:


      Also, the SNPs themselves:


      There have been some R*s (RM207) found, very unusual, mostly around Pakistan. R1* (R-M173) is also rare but has been found scattered around Iran. There is also an unusual R1a* (R-M420) named Pickering, who has tested both STRs and SNPs, as well as at 23andme. R1a1* is also quite unusual, seeming most at home around Armenia. The bulk of what we think of as R1a has been in what until very recently we called R-M17/M198.

      About a month ago, someone with a DYS392 value of 13 (it's usually 11 in R1a1a) tested positive for R-M198 and M-17, and negative for M417 and rs34297606 aka PAGES00007. This means the bulk of all previous R-M198/R-M17s are now R-M417/Page007, one level downstream from R-M198/R-M17. This includes all Ashkenazi R1a1a I know of.

      The main STR cluster for Ashkenazi R-M417 tends to be quite close to Ysearch user FCUFG. You can see several additional markers, which may or may not be modal, from my kit, user CXD4T. This subcluster is by no means small, but is fairly close/tight and members of it largely identify as Levite. A few of the members of this cluster claim Sephardi ancestry from Iberia with names like Levi.

      I anticipate seeing a small number of Sephardi and Mizrahi R-M417s who will mosre closely match the R-M417s from places like Oman and Qatar, which while not common are not rare either. Unfortunately Mizrahim who have tested to 67 markers _are_ incredibly rare.

      This Ashkenazi cluster is best categorized as a subgroup of what Peter Gwozdz calls the K ("Kurgan") cluster. I'll stay away from the debate over whether this term is historically apt, the cluster itself is real enough, and its modal is essentially the R1a1a modal, in other words Ashkenazi R1a1a forms a close-knit subcluster of the earliest known form of R-M417, not differing much from Pakistani or Kazakh or even some Isles R-M417s. Unfortunately, in those areas R-M417 is most interesting, few kits have tested more than 12 markers.

      A recent Ashkenazi R-M417 kit in a Walk Through the Y revealed no new SNPs beyond M-417. A non-Ashkenazi M417 of the K cluster but not the Ashkenazi subcluster of it with origins from Anatolia has also tested negative for downstream SNPs in a WTY, and positive for all SNPs up to M417. I highly encourage Q's to do WTY if you can get the resources together and identify the best candidate(s). Sometimes finding nothing is as useful as finding something.

      Another exciting development in the past year was the discovery by Peter Underhill of the SNP M458, which is for the moment called ISOGG R1a1a7, but which will be renamed once the M417 level is added. This has not yet been found in any Ashkenazi R1a's I know of but is found in a significant number of Poles, eastern Germans, western Russians, and other Eastern Europeans, mainly of Slavic origins or in areas with large historical Slavic populations. There is a further subclade, very rare, of M458 currently called R1a1a7a defined by M334.

      An additional SNP downstream of R-M417 has been found in a McDonald, who has what was called the "Somerled motif." This is a Scottish subgroup of the one of two principle Norwegian clusters that has a YCAIIa,b of 19,21, rather than the R-M198, K, and Ashkenazi default of 19,23. It is not expected to be of any use for Ashkenazi R-M417s.

      I have no vested interest in promoting any particular place of origin for the Ashkenazi subclade, and moreover don't feel qualified to. I believe there is at this point insufficient evidence to determine whether the Ashkenazi M-417 cluster arose in the ancient Levant proper, or somewhere around the Caucasus or Steppe. It may have entered the Ashkenazi gene pool from an Indo-European people from the long sojourn of Jews in Persia, or from something like Kurdish or Armenian converts to Judaism (I believe that some of this occurred has been documented), or from Khazars, as Ellen Coffman speculated <http://www.jogg.info/11/coffman.htm>. Equally, it may have been in the Holy Land before 70 AD since this area was a longtime crossroads, and would never have been genetically uniform.

      Put another way, I believe the dating and distribution of Ashkenazi M417 and upstream subclades could be consistent either with gradual migrations of prehistoric peoples from a glacial refugium in Central Asia to the Near East by biblical times, or with a more recent origin from somewhere north of the Near East, i.e. between Eastern Europe and Central Asia.

      I'd be interested to hear from the Q experts as to their thoughts on Ashkenazi Q1b since the two cases do seem at least to me to be parallel.

      Finally, I encourage all males who have tested at 23andme, if they have not yet done so, to contribute their yDNA results to Adriano Squecco's spreadsheet. You can get an idea of the value of the results here:


      You can see here that +L49 has been found in some R1a's, but it's still unclear (at least to me) how this fits in. +L69, usually used in R1b1b2, has also been found in R1a but this marker is palindromic and "flippy," so probably best ignored.

      If you think I've left anything out or been unclear please feel free to follow up by CCing me when replying to the group, or by inquiring to my email off-list.

      All the best,

      A.J. Levin
    Your message has been successfully submitted and would be delivered to recipients shortly.