These are the kinds of graphics I was referring to in my first post on how to create interactive charts with Google Docs. They're based off Davidski's work in the Eurogenes Genetic Ancestry Project, which now includes DNA from one of the kits I manage.
I've now had DNA samples from family members tested with both Family Tree DNA (FTDNA) and 23andMe, which each has their advantages and disadvantages. (I'm going to post more on this subject in the future. For now I'll suffice to say, if you're interested in genetic testing for genealogical purposes, FTDNA is the way to go.) The raw data files from FTDNA's Family Finder test, and 23andMe's only product offering, can be downloaded and used for this type of analysis. Although I've played around with the genetic software Davidski used to produce this data, my understanding is still too limited to make my own charts from scratch. (The file format from each company use a slightly different format, so I've only been able to use 23andMe data in my experiments thus far.) I merely used the data in his spreadsheet to create these charts, including only the populations I'm interested in.
I'm not even going to attempt to explain what these charts really mean, except to say the software infers genetic ancestry based on DNA similarity. The samples shown in the charts are identical, but each of them was produced by telling the software to divide the dataset into K ancestral populations.