Connecting to Communities

 

The news from Ancestry.com is that they are once again adding to their website more of what they dub "DNA Communities." In particular—and this is what caught my eye—the latest addition includes an increase from seventy one to over two hundred communities affiliated with the southern United States.

This is not the first time Ancestry has added to their collection of DNA Communities. About this time last year, they put out an announcement detailing fifty five new communities, mostly surrounding the Mediterranean, plus their first east African community. As they did last year, I suspect the new communities will appear gradually as they roll out the updates to subscribers.

The story behind how Ancestry puts together these communities is interesting in itself. Using a network analysis method known as community detection, Ancestry begins by assembling a network of the millions of subscribers in their database. From that point, they cull out the individuals who share a greater number of DNA matches with each other, and separate them from the rest of the network. They then fine-tune their results with information from subscribers' family trees.

Of course, they keep an eye on details like migration pathways. One particular clue is the observation of differences between birth locations of parents and their children. But there are also factors which determine strength or weakness of such community connections. For instance, the greater the number of generations separating the subscriber from a community, the less likely it will be for a connection to be identified.

In other words, I would not likely find myself part of my paternal grandfather's New York Polish community (if there was one) because of several details. First, the rest of my grandfather's community immigrated to Milwaukee, not New York, so he would be an outlier. In addition, any connection would be to my grandfather, not a closer generation—and, for that matter, those were both "long" generations (spanning nearly fifty years in one case). 

So, how did my mother's southern genes fare in this latest update? As soon as I saw the news about this update, you can be sure I checked. I'm hoping this case, too, will be a matter of rolling out the results gradually. My results, at this point, haven't budged from where they stood before the announcement: only two communities, including one sub-community. My DNA Communities: Georgia and Florida Settlers, with its sub-category of South-Central Georgia, Lower South Carolina, and Florida settlers; and secondly, early Georgia Coastal Plain and Northern Florida Settlers.

Of course, I'm hoping to see an update, but would not be surprised if the results remain the same. These are, after all, my ancestors who are portrayed by these results. And that was just exactly who they were: just as the DNA Communities portrayed them.