Genetic Communities: A quick glance at Ancestry.com’s latest feature

Genetic Communities: A quick glance at Ancestry.com’s latest feature

The verdict? This could turn into a very useful tool. More and more Ancestry is guiding us towards bringing out more of our ancestor’s stories, as opposed to just standard genealogy, and this is a successful part of that effort.

I noticed this morning that my DNA Homepage had been updated…and a new feature had been added (in Beta): Genetic Communities.

 

Capture10

Under what’s now termed Genetic Ancestry and in addition the Ethnicity Estimate, there’s now a section called Genetic Communities. 

Capture1

When you click through to Genetic Communities you’re presented with the communities that Ancestry has calculated that you, and member matches that share ancestors with you, belong to. The information presented can be a bit general, but it does provide pretty accurate/helpful historical context. For me, I was presented with 2 “Possible” communities that I could belong to:

Capture2

In clicking through the Settlers of Massachusetts, Vermont, Rhode Island & Connecticut community, I saw some pretty cool graphics:

Capture3

The migration lines are animated, and you can see some of the stories that support the idea of a community on the left side pane. If you click a time frame in the left pane, you’ll get a small write-up on what that time encompassed as well as showing how many of your relatives are potentially covered in that description. Additionally, you’re shown graphically where your connected Member Trees have ancestors during that time.

Capture4

If you click on the list of users from your family tree, you’ll see the list of your ancestors who lived in the covered area, during the covered time.

Capture12

As you scroll through each time span, you’ll be able to see a representation of how your ancestors, and some of your connected relatives, migrated as time went on. You’ll also get more information about the events and context surround those time periods, as well as a complete list of your relatives who match this classification.

Capture11

If you go back to the main page for your community, there’s an option to show “Connection” and clicking that will give you more information on your family tree:

Capture13

I clicked Connection and it details how sure Ancestry.com is that I fit an identified community:

Capture5

On the lower part of the screen, you can see they have grouped your DNA matches between you and the broader community, as well as the last names associated with your lineage.

Capture6

There is definitely a LOT to click through and enjoy. For instance, each time period has a listing of “Historical Insights” at the bottom that will help give you broader information about what was occurring during that period.

The verdict? This could turn into a very useful tool. More and more Ancestry is guiding us towards bringing out more of our ancestor’s stories, as opposed to standard genealogy, this is a part of that effort.

My family line is heavy on English settlers to Massachusetts in the early-to-mid 1650’s, and Genetic Communities identified that successfully, but since I know this era relatively well it was easy picking out some less than accurate facts that were attributed to my family. Ironically, the tests from Felice’s Grandmother were much more accurate and helpful.

Capture8

It’s not the case usually that any area of African American genealogy is more complete than white genealogy, however Felice’s grandmother quickly showed a “Very Likely” match to the African Americans in North Carolina community.

Capture14

As you can see, her community is pretty well established and is Very Likely to return matching records. Reviewing the notes that go along with this community was fascinating, and will likely help my work to further her line. I didn’t understand the prevalence of the West Africa/North Carolina flow of slaves, and it helps explain why so many references to North Carolina appear when tracing her various relatives, even though her family has very few ancestors in the state from 1850 onward.

Just like with my line, however, the general nature of the historical notes is pretty clear as the migration patterns and experiences of this group in general gives way to what we know about her family in fact.

So, my first impressions are that this is major new feature that delivers very interesting information, and helpful context and general information about both my lines and the lines of those related to us, but it likely will need further refining to bring it’s true value into clearer focus.

Ben Franklin, and how We’re Related took me from chasing my white whale, to chasing my tail

Ben Franklin, and how We’re Related took me from chasing my white whale, to chasing my tail

If you haven’t used the “We’re Related” app from Ancestry, you should give it a try. It seems a little silly to begin with, but there’s some value to it (a previous post: We’re Related app is a lot less frivolous than it first appears). But a recent tip from the app seemed to lead me to breaking down a major brick wall, before I realized I was fooled by my own published research.

First, a little background on my brick wall. The Tradewell (it was actually Treadwell prior to ~1840) line is one of the trunks of my family tree growing up. We all have those, the handful of family lines that dominate our history and identity when we’re growing up, despite having 8 Great-Grandparents, the Tradewell line was one I knew a lot about growing up. Its also the line that has a lot of genealogical significance, because my Great-Grandmother Myra (Tradewell) Morse submitted her first DAR application in 1907 and that not only gives me a lot of information on this line, subsequent cousins and relatives who applied attached sworn copies of family bibles to their applications, which preserved a great deal of info that otherwise would be long lost to history.

But Myra’s line ends with her Grandfather, and he is only one of two 3x GGP’s in my line that end so early. James Bennet Tradewell died in my hometown of Racine, WI in 1885, moving here just after 1840 from Schoharie County, NY. The 1830 Census includes an older Rueben Treadwell living in Schohaire County along with two younger Treadwells, Ephriam and James B. In 1840, Rueben is no longer listed but James and and Ephriam are. By 1850, from my research and from census records, there are two men in South East Wisconsin named James and Ephriam Tradewell who were born in New York, and those two names are also not present in New York for the 1850 Census. So even making a guess that Rueben is James and Ephriam’s father, and that James and Ephriam moved to Wisconsin Territory around 1843, I can go no further. I have very little evidence of even those links, and the line beyond Reuben goes dead.

I’ve also done research on the broader Treadwell line, which is largely threaded through Thomas and Edward Treadwell who arrived from England in the 1630’s and their decedents settled mostly in Connecticut and New York. My current working theory is that Jacob Treadwell and Hannah (Trowbridge) Treadwell are the parents of Reuben, but I have no proof whatsoever that this is true.

imgonline-com-ua-twotoone-vkoHNUp2D3tV

I created a “Working/Uncertified” for the Tradewell tree, to match these two half into a theoretical whole that I hoped would spawn some “shakey leaf” hints on Ancestry.com, and to further my research. Unfortunately, that didn’t work. I still have no clue as to the names of Jacob and Hannah’s children, nor who Reuben’s parents might be, so it’s one big brick wall.

Until, as I was waiting in line at the grocery store and scrolled through my latest We’re Related matches and I see Ben Franklin. When I expanded the link, I almost fell over when I saw Hannah Trowbridge connected to Reuben Tradewell who was in-turn connected to James Bennet Tradewell! I was stunned…this app that links me mostly to celebrities I couldn’t care less about had just broken down one of my top-5 brick walls!!!

Screenshot_20170323-200000

I was giddy as I got home, made dinner, begrudgingly hung out with my family, until I was finally free to do a little research. I did a search on Public Trees for Hannah Trowbridge, with her husband’s last name set to Treadwell and I hit pay dirt! Right away I see that all of the children I’d guess were theirs were also listed. I looked at the attached facts, and didn’t see anything that linked the children to the parents, so I decided to review the Ancestry Family Trees to see if the trees that were sourcing this tree had more facts than this.

And that’s when I realized I wasn’t harpooning my white whale today, and wasn’t related to Benjamin Franklin. My Working/Uncertified tree was the source of the link!

Match

The We’re Related app had built a link between myself and our common ancestor using the data available, and it turns out that my test tree was the only link. And since the algorithm doesn’t read the titles of trees, or the description that I attached to the tree, it treated it as a valid, proved link.

I’m starting to re-think my strategy of keeping my Working/Uncertified trees Public. I’m a big supporter of keeping our trees Public, but in this case not only did my strategy fool the app, there are several user trees that now have this link shared as a fact. As time goes on, and these electronic records are shared, and re-shared, and memorialized outside of Ancestry.com, I’m afraid I’ve introduced misinformation into this family line.

We’re Related app is a lot less frivolous than it first appears

We’re Related app is a lot less frivolous than it first appears

I’m not sure why I first downloaded We’re Related, but it felt rather silly. Almost immediately the app is telling me I’m related to Madonna, Johnny Depp, and Bill Clinton which seemed like a gimmick. And, let’s be honest, it comes off more as a marketing tool than genealogy tool.

But to discount this app is to be missing out on some decent family history research opportunities as well as missing what could be the dawn of a great new tool in genealogy!

If this technology is ever leveraged against some of my brick walls instead a gimmick like linking me to Blake Shelton, Ancestry might really be on to something.

The feature that I’ve found most useful, and that I wish was more available in Ancestry.com, is the projection of relationships beyond what you’ve identified in your tree. The tool collects data from other sources and makes an educated guess about your direct ancestors and presents them as a line that happens to match the line of a celebrity. This conjecture is like a “shaky leaf” hints taken to the next step. Instead of using the data you’ve entered that matches either what another Ancestry member has entered, or a fact match that shares the same information as you’ve entered into your tree, We’re Related will build out your tree using your information as a starting point, and try and build a map that uses the best available data to connect your line to another. The potential power of a tool like this, when properly executed, is what excites me…much more so than potentially sharing ancestors with DJ Spider One.

Using my match with John McCain as an example, let me show you what I mean…

Screenshot_20170318-235822

This is the first match I decided to trace and see if I could prove out the connection, and given that I have several DNA lines that are proved that are 7th Cousins or above, it seemed like I’d have a good chance of mapping it out.

Screenshot_20170319-002851

As I expected, I’m several generations older than Sen. McCain so I had more ancestors to prove. Clicking the link showed that James Morgan was on my Father’s, Mother’s, Father’s line which is well documented. I wasn’t aware of the Morgan name previous to this, but when I started tracing back I quickly recognized my 5xGGM Mary Goodwin.

Screenshot_20170319-140706

From here I just followed the shaking leaf hints back to James Morgan. My main tree is a Working/Uncertified tree so I’m pretty liberal on the sources I accept to establish facts in the tree. I stay away from Ancestry member trees, but I’m comfortable using secondary or greater sources like those in the Lucius Barnes Barbour Collection (a collection of records as transcribed by Lucius Barnes Barbour in the mid-1800’s) and the Family Data Collection (genealogical data gathered for scientific study that was not properly documented, and set to be destroyed, before it was saved and entered into an electronic database) despite the risk that the data isn’t very accurate. It’s not unusual for there to be transposition of dates, or years like 1703/4 trimmed to 1703, but by and large the overall information is accurate, if not enough to be considered genealogical “proof”.

Screenshot_20170319-002939

I was able to establish the link from Mary Goodwin to James Mason just following “shaky leaf” hints, and this demonstrates how this could be a powerful genealogical technology. Unlike the shaky leaf hints, which are based on the top 10% of most used sources, and are closely tied to information you’ve already entered on a relative, We’re Related is making suppositions based (apparently) on an algorithm that can draw the line between what you know, and what it guesses is true, to build a potential line for you.

In this case, the system’s supposition was accurate, and it helped me build a link that I hadn’t built. I have other examples, which I’ll write about later, where those suppositions weren’t accurate. But if this technology is ever leveraged against some of my brick walls instead a gimmick like linking me to Blake Shelton, Ancestry might really be on to something.

Until then, We’re Related is more than just a gimmick, and if you get lucky, you might accidental solve a family mystery.

(UPDATE: Ancestry has incorporated features of “We’re Related” into the main site…and it’s not what we’d hoped Ancestry.com takes another step away from it’s genealogical roots…)