No announcement yet.

what does really mean "Iberia" and "British Isles" ?

  • Filter
  • Time
  • Show
Clear All
new posts

  • what does really mean "Iberia" and "British Isles" ?

    Hi everybody,
    I'm french and my DNA results show "Iberia" and "British Isles" origins. But it seems to be very generic terms. Does it mean one of my ancestor was spanish/portuguese or, more probably, the "Iberia" term also covers some french populations (like the Arverns in central France) ?
    With the "British Isles" term, it's even more ambiguous, does it mean Celtic ? Saxon ?...
    I'm a bit lost ...
    If someone could help me, it would be great !
    Thank you

  • #2
    The way these labels originate seems to be: First, some sort of statistical analysis detects a large cluster of "reference samples" that are more similar to each other than they are to other clusters from different geographical areas. After some refinements (AncestryDNA has posted some documentation about how they select the samples for the final "reference group", and I think the process at FTDNA must be similar), it is now necessary to find a label for the cluster. However, the boundaries of the cluster (that is, the apparent geographical origins of the people who whose samples are included in a given cluster) are often not well defined (except statistically), so the labels become to some extent arbitrary. In other words, your results show you are more similar to the clusters that are labeled "Iberia" and "British Isles" than you are to any other clusters that are presently well characterized, but that does not mean that you are either "Iberian" or "British" in any meaningful sense.

    Missing from this discussion, obviously, is a meaningful "French" cluster! The difficult legal status of operating a DNA testing business in France is one reason for this. There are many historical reasons to expect genetic similarities between France and the British Isles, and also, perhaps to a lesser extent, between France and Spain. Many of us wish for a serious attempt to obtain a much larger set of samples from all parts of France, from people whose ancestry back at least to the 16th Century is known. However, it should be remembered that the countries of Europe are quite small, and that there has been extensive migration within Europe in historical times. The genetic differences between French people and their neighbors are not likely to be particularly large or well marked.


    • #3
      When you are viewing your myOrigins map, you can click on each of the population cluster names, or their colored areas on the map (British Isles, Iberia, etc.), and it will show a description in the box at bottom left (which has two tabs: "Shared Origins" and "My Ancestral History"). These descriptions are probably just like the ones shown on the FTDNA Learning Center page for Population Clusters in myOrigins. There is also a myOrigins Methodology Whitepaper explaining the methods and results used for myOrigins, that you might want to read.

      But these descriptions really don't seem to answer your questions. About the only part in those population cluster descriptions that might show how France is included is the last two sentences of the British Isles cluster description:
      Normandy later invaded and solidified cultural and economic connections between the British Isles and continental Europe. To this day, these ancient occupations and trading practices left a lasting impression on the genetic relatedness between populations in the British Isles cluster and Southeast Europe, Scandinavia, and West and Central Europe clusters.
      For the Iberian cluster description, there is this:
      Furthermore, this cluster shares a common genetic heritage with parts of Western and Central Europe, the British Isles, and Southeast Europe due to the settlement of Celtic and later, Germanic tribes
      The "West and Central Europe" cluster includes France. It is possible that some population clusters also cover certain populations in other clusters, but I don't know enough about it to say that with certainty.

      Overall, though, you really shouldn't put too much weight on what myOrigins predicts. Rather, look at your DNA matches, to see where they come from. Your matches are a better indicator for your recent (up to 500 years) ancestry. I believe that myOrigins goes further back than 500 years, usually beyond what we can find out about our ancestors in records.


      • #4
        Unfortunately, there is no guarantee that any admixture algorithm (including "myOrigins") can tell you anything about where your ancestors might have been 500 years ago. Rather, the algorithms measure your genetic similarity to an arbitrary set of "reference groups" consisting of DNA samples from modern (and presumably still living) people whose ancestry is believed to come from specific geographical areas, and who are genetically more similar to people within their "reference group" than they are to people from other "reference groups". If you dig deeply enough into the literature describing the development of the admixture algorithms -- and that's not an easy task! -- you will eventually find a statement to the effect that the algorithms use modern DNA samples as proxies for our deceased ancestors, whose DNA samples are not available to us. The statement that the "reference groups" or the admixture results pertain to some reality 500 years ago, or any other figure, is not based on evidence!


        • #5
          Thank you everybody for your answers.
          As you say, the problem is certainly the lack of french data ...
          But anyway, even if the British Isles are rather small, there is a big difference between a celt and a saxon ! Too bad the results could not be more refined ...


          • #6
            While there may be a big difference in terms of history, culture, etc. for Celts versus Saxons, I'm not sure it's clear yet whether there was a significant genetic difference between them, and if so, whether it could be detected by the arrays of SNP's that are currently used for commercial autosomal DNA testing. This is a question that may eventually be answered from "ancient" DNA. There is also the question of whether those differences, if they existed in "ancient" times, have persisted in any modern populations. I hope we will eventually have a clearer understanding of these questions!