OK, so we all know that those home "test your dna for diseases" kits are pretty much a sham according to various government investigations and sting operations.
That's fine with me, because I'm not really interested in testing my dna for diseases. (Well, I am, but only if it's really accurate, which it isn't, so thanks I'll keep my money).
I AM interested in DNA testing to investigate my ancestors/heritage. Specifically, I'm curious if I carry any of the DNA that is considered a dead give-away for having Jewish ancestry. Some of my ancestors were from eastern Europe and I've had quite a few people over my life either ask me if I'm Jewish (once I was asked if I was Palestinian ...) or assume that I have Jewish ancestry.
I know that quite a bit of the ancestry DNA stuff is either dodgy or just incredibly imprecise. But I've always had the sense that in the case of, say, Ashkenazi Jews, genetic tests can actually be fairly certain (though of course it's possible to have Jewish ancestors whose markers haven't been identified as Jewish yet, etc.).
How would a scientist like yourself go about deciding on whether DNA tests were accurate enough to detect if a person was Ashkenazi (that would probably be my background if I have Jewish ancestors) and how would you manage to pick a testing company that wasn't just peddling BS or going to to try to sell you products for conducting research on your ancestors, etc.?
1. There's nothing wrong with the DNA tests for unifactorial or oligofactorial diseases. If you're carrying a common mutation, or at least a mutation that's tested for, it will pick it up. The problem is when they use a bunch of markers that either aren't specific or well-studied, and they start playing the correlation game with them. That's probably the controversy you're talking about. But for stuff like Tay-Sach's, Gaucher's, and Niemann-Pick...the tests are going to work as well as they possibly can (that is...they won't pick up the 1-in-a-million novel mutation, but you're more concerned about the 1/7-1/8 carrier frequency with something like Gaucher's).
2. It would all depend on what markers they're using, how many, the database of "known" genotypes they're comparing yours to, and whether they're testing autosomal, Y-chromosomal, or mitochondrial DNA. The more markers, the better...the more knowns, the better. Autosomal DNA isn't really useful for anything beyond a generation or two...at that point probably all you'd be able to say with relative certainty is a region (e.g. "northern Europe"). Y-DNA is a little better, but it can only tell you if you recently shared common ancestor with a known. Mitochondrial DNA testing is pretty solid, but only if you're a near-perfect match for a known. If you're a perfect match, that's great...but if your mtDNA is off by even a little, mitochondrial DNA acquires mutations at such a slow rate that even a minor variation from a near-match could put you off by tens of thousands of years, using the "molecular clock" method.
That being said, if you're specifically looking for Ashkenazi Jewish ancestry, there probably isn't a better-studied group, as far as genetic testing is concerned.
