Thank you for that link.
I took a look at the European specific PCA's for all five ancient European samples. (four from ancient Bulgaria, and one from ancient Denmark)
They can be found on page six of the Supplementary Data section here:
http://download.cell.com/AJHG/mmcs/j...0459X.mmc1.pdf
I think they have to be used with caution. In the case of the Danish sample, it rather contradicts the post capture global PCA plot, as in the global plot, the Danish sample is near Finland...here, the Danish sample plots right at the beginning of the "Southern European" section.
In terms of the ancient Bulgarian samples, it's difficult to draw conclusions; I don't see a definitive narrative, either in terms of age or culture or class.
1) Sample P192-1, a Thracian culture sample found in a pit sanctuary, 800-500 B.C. and the mtDNA is U3b
2) Sample T2G2, a Thracian culture sample dated to virtually the same time frame (850-700 B.C.) and found in a tumulous burial. The mtDNA is HV.
3) Sample K8, a slightly younger (450-400 B.C.) Iron Age sample, also found in a tumulous burial. No mtDNA determination could be made.
4) Sample V2, the sample which appears in the global plot, is one thousand years older and dates to the Late Bronze Age (1500-1100 B.C.). It is an inhumation, flat burial. No mtDNA determination could be made.
Looking at the European specific PCA's, the P-192-1 Thracian Iron Age sample lands in the Tuscan plot.
The plot for the T2G2 Thracian Iron Age sample lacks a lot of definition, probably partly because they were unable to extract as many snps as they did from the other remains, but he seems to land right where the Tuscan, GB and CEU samples meet.
V2, which is the much older Bronze Age sample shown in the global PCA plot, is a little further away, but is right next to where the Southern European clusters begin. This would correlate pretty well with the probably more accurate V2 global plot, where, if you increase the size of the image, you can see that he plots right over GB, Iberian and Tuscan samples.
The K8 sample is the only one that doesn't fit the overall Southern European "look" of the other samples. It seems more "Northeastern" and plots in the middle of the GB and CEU samples.
I was expecting the Bronze Age sample to perhaps be more "North-Eastern" in orientation, but that doesn't seem to be the case. If I were going to guess, perhaps the K8 sample is from an intrusive, later Iron Age Culture...there's a difference of about 300 years if we stretch the dates a little.
Or perhaps it's down to geographical sub-structuring and these attempts to pin a specific yDNA and mtDNA to certain specific cultures is not going to be informative at this relatively late date in European history. ( P192-1 and T2g2 are in the south, V2 is in the northwest, and K8 seems to be from the central eastern portion of Bulgaria.)
Also, as I mentioned above, I don't know how seriously to take these European specific PCA's, as the Danish sample is as close to the Tuscan cluster as some of the ancient Bulgarians.
The global PCA plot in the main body of the text (Table 3) is probably more accurate, and that plots the V2 Bronze Age sample right over on top of what looks like the Iberian samples, some Tuscan samples, and some GB samples. I wish they had provided global plots for all of the ancient Bulgarian samples.
See:
http://images.cell.com/images/EdImag...G/ajhg1537.pdf