01-12-2014, 03:04 PM
sekker wrote:
[quote=PeterB]
Just for the record, as a geneticist, I don't see how anyone can claim to be "99.5%" anything. That's a problem with their science, right there. Sure, some alleles might be more common in some populations, but I think you'd be hard-pressed to be able to accurately define things in that way.
Peter, there could be some recombination between haplotypes to get 99.5%. At that low percentage, all it would take is a single chromosome exchange on one arm.
No, monogamy doesn't matter in this case. You personally are from a single lineage.
Whether you are biologically related to anyone in your family, well that's another story altogether...
I think the issues with regard to parentage are some of the most critical that 23andme are not dealing with properly. In some march of dimes birth defect studies - these are opt-in studies - as much as 15% of paternity is not confirmed by the DNA tests... My recommendation is to test you and your mom; and ask her honestly whether testing Dad is going to be an issue,,,
I take issue with the idea that there are absolutely nationality- or ethnically-specific haplotypes. So it's not about recombination or anything of that kind. Sure, you might be able to make an estimate, but an estimate is NOT the same as a percentage, and I think it's highly misleading to say that anyone is some "percentage" of anything-- in the same way that it's misleading to say that your dog is 70% cocker spaniel, 20% Doberman, and 10% poodle (I just made that one up!).
I think 23andme has messed a LOT of things up. It's not just about their misleading medical claims, either.
Estimates of nonpaternity vary from study to study, but they've sometime reached the 30% mark.