Please help me interpret my 23andMe ancestry results

I’m reading through that, and I here are some thoughts as I come across them:

[ul]
[li]Genetics is bad, because racists and Nazis. Yeah, racists existed long before genetics, and just because racists occasionally picked up on genetics as a tool to administer their hate, doesn’t mean genetics is worthless. Eugenics was terrible, and if anything modern genetics can explain how wrong headed it was in even achieving its disgusting objectives.[/li][li]Populations aren’t a thing. OK, if there isn’t such a thing as a population/group/tribe, then there also isn’t such a thing as a family. People are all 99.9% genetically identical anyway, so relatives and strangers are interchangeable. As I’ve explained, this kind of clustering is an emergent property of genetics. Hey look, we’ve got a bunch of people that aren’t just related 99.9%, but 99.95% (or whatever) and they all live in the same region, lets call that a population/group/tribe/race/whatever. And look, when one of those people goes and lives someplace else and has kids, we can see that those kids are related to that original group by 99.9125% (or whatever), and that’s way more than chance. Now if we run across somebody that is related to that group by 99.9125% we can say that they have some ancestors from the region that group lives/lived in.[/li][li]Genetics can’t fill in all of the details about the movements of people, therefore migration is a bogus concept. Genetics never claimed to give the details of exactly how people from one area moved to another area. All it can do is show that due to genetic relationships this group over here is related to that group over there, and by looking at mutations and other data, that at some point people from the original group moved and started a new group. For example, if group F1 has the unique sequence AAAAA and group F2A is ABAAA and group F2B is AAABA, then the simple explanation is that people from F1 founded both F2A and F2B. So don’t use the word migration to tell this story. OK, groups F2A and F2B magically sprang from the earth, but just happen to look a lot like F1.[/li][li]Admixture doesn’t tell us why people are having kids with their neighbors. So what? Genetics don’t record the why of mating, just the product of it. Hey, we found some ancient DNA and it looks like a mix of a few different groups, not just like one group from someplace else. Great, now we know that those groups intermingled. Do we know the extent of it? Maybe not, perhaps it is just chance we found the one intermingled family in the entire region, but as a non-archaeologist, it doesn’t seem any sillier to me than telling a whole story based on a single refuse pit.[/li][li]Telling non-European people about their history is just another form of colonialism. I get mistrust of European and Western motives. We’ve sure done some terrible things to lots of indigenous people when we met them. Obviously I have white privilege, and I guess it leaves me thinking that more knowledge is better than no knowledge, and that collecting some DNA and writing a few papers isn’t exploitative in the way that say, stripping them of natural resources is exploitative.[/li][li]We’re more alike than different. This is the same argument as before, humans don’t come in different kinds, which is just silly. We don’t come in different kinds as far as who should get human rights and treated with respect. We come in different kinds as far as how that last 0.1% of DNA affects the differences that we do see.[/li][/ul]
That whole paper reads to me like the author saw Reich’s book which said genetic data suggests Polynesians and other Pacific Islanders descended from a group that lives/lived in Taiwan. But he knows that the story isn’t that simple, so he jumps up and down and claims that genetics is pointless, except for finding the cause of diseases and stuff.

Resurrecting this not too old thread with an interesting news article. A pair of identical twins each sent their DNA off to 23andMe, Ancestry.com, and MyHeritage. As expected, their ancestry breakdown varies company to company, but it also varied twin to twin within company. TL/DR Don’t get hung up on exact percentages. Buy another kit and spit in another tube, and you’re likely to get slightly different results. If you don’t like your results, use another company. There is no good way to know which company is correct.

The twins’ raw genotypes are 99.6% identical. That last 0.4% of difference is almost certainly error, as this type of genotyping is not sensitive enough to detect the very very small differences that exist between identical twins. A mutation would have to be exactly in one of the SNPs that is being tested, and ~2500 mutations in just the right spots is impossible. So, genotyping error.

The point of the article is that one twins is, for example, 25% Eastern European, and the other is 28% Eastern European. About 2500 SNPs changed, and that was enough to push results around by a few percentage points.

The different companies are giving different answers because they use different reference panels and possibly different methodology to create estimates. According to 23andMe and Ancestry.com the twins are both mostly Italian and Eastern European, with some Greek/Balkan. MyHeritage says they are almost completely Greek/Balkan, and hardly any Italian or other Eastern European at all. I would not have expected to see such large differences, but I also do not know the strengths and weaknesses of the various reference panels.