http://bga101.blogspot.com/2011/01/amer … altic.html

Amerindian-like sequences in Baltic Finns (aka. phased data is the way forward)

I thought this Finnish experiment was going to be a real challenge, and something of a long-term project. In fact, I managed to get it all done in a couple of hours. And it turned out great too.

Basically, I wanted to see if I could take a more detailed look at genome-wide sites of possible non-West Eurasian admixture in some of my Northern European reference samples. Up to this point, I’ve been studying segments containing mixed up sequences of alleles from both parents, usually with only one coming from a potentially exotic ancestor. Thus, there was often some guess work involved, trying to figure out what the picture would look like if the accompanying sequence, the one of European origin, wasn’t interfering with the results.

The only way to get around this issue is to use phased data, in which the alleles are sorted as coming from one parent or the other. This makes it easier to compare segments, and run all sorts of statistical analyses to check their biogeographical character. Indeed, with phased data it’s even possible to make two people out of one, and run them at the same time, just to see how different each of their parents were. That’s what I’ve done here with some of my Finnish samples, except instead of comparing whole genomes, I ran MDS plots of data from segments that showed relatively high affinity to East Eurasia.

I expected to get results where one sequence clustered amongst Siberian and Central East Asian samples, and the other near the Europeans. And yeah, I got that, as the two plots below clearly show (red underline = sequence of putatively non-European origin, blue underline = European).

Pretty cool stuff, but nothing really extraordinary. After all, Siberian genome-wide influence in Finland is well documented, and the latest consensus is that it got there during the Bronze Age, along with Uralic warriors and traders from the Volga-Ural region. But here’s something I wasn’t expecting…


The above two plots show that Baltic Finns carry sequences that are most similar to those of Amerindian populations, and not because of a lack of reference samples from across Siberia. These sequences don’t appear to be all that unusual in my Finnish samples. What this means is that we’re not just looking at noise here, but genuine links between Finns and Amerindians. I have no idea how old this stuff is, but probably really old. I suspect it got to Europe with the aforementioned Uralic migrants of the Bronze Age, but I’m not really sure. I doubt Eskimos ever made it to Finland.

The other interesting thing is that, usually, when I’m searching for non-European segments in one sample, I pick them up in some of the other individuals present in the analysis. Note some of the Chuvash and North Russians streaming towards East Asia on the plots above – they most likely have Siberian sequences in the areas tested. However, I’m yet to see any native Europeans other than Finns showing the unusually strong links to the Americas I identified here, even when checking small areas of the genome of around 5 megabases.

I haven’t tested Germanic Scandinavians like this yet, but soon will, and I have a feeling they too will score some hits in the Americas, due to their results in some of my ADMIXTURE runs. In any case, it’d be nice if someone could try to corroborate these results, and expand on them, in a formal study. We could learn a lot about European population structure and history from detailed extended haplotype analyses.