Check Your Sides: Why Ancestry ThruLines Need Verification

I've mentioned this before to several of my genealogy friends but wanted to blog about it to remind folks that ThruLines are HINTS.  Hints are suggestions which you can verify or prove wrong....or even totally ignore-- that's your choice.

This ThruLines match of mine (shown above in split screen of ThruLines beside Profile page) is a great example. This is not an error by Ancestry. In the pop-out panel of the ThruLines, it shows the match as paternal because if the ThruLines is correct, the match would be through a paternal line. However, the profile page shows that the match is on the side that I (or Ancestry) has determined to be the maternal side. My mom has tested and the match shows she matches them for the same amount. In situations such as these, keep in mind that you could still be kin to a match through that path and share no DNA. Still you would have to verify that connection.

I wish Ancestry would do a comparison of the two sides in their code and kick out any ThruLines where the ThruLines hint couldn't possibly be true. I don't guess that will happen. Of course, this would not help with the ThruLines where the side of connection is correct but the path isn't. Pay attention and verify always, especially if you are going to be basing any further research or theories on the connection.

Ancestry’s Updated ThruLines: What I Like and What I'd Like To See In Future Updates

The updated layout of Thrulines appeared in my account yesterday.  I think it's a great visualization tool.  It lets you see how many DNA matches have been identified AND fall within the specifications to show up on Thrulines for each branch as you can see in the image below of my Thrulines out to my 2nd Great Grandparents.  

There is nothing that checks to see if the matches are on the correct side though and even if you do evaluate and add or choose not to add there is no option to mark a match as "not valid thru this line."  Case in point, my paternal 2nd cousin once removed (also paternal for him) who is also a maternal 4th cousin once removed (thru his maternal lines) but shares no DNA thru the maternal connection. He shows up in the Thrulines of the Paternal line and the Maternal line. While we are close enough thru the maternal line to have many shared maternal cousins I know he doesn't share DNA with me thru that side because he doesn't share any DNA  with my Mom who has also tested at Ancestry.

Below is a look at my Paternal Grandmother's branch.  The Thrulines doesn't offer the next generations for my 3rd Great Grandmother, Margaret Frizzell. This is due to pedigree collapse as she is a sibling of James M Frizzell my 3rd Great Grandfather(they are the children of Nathan Frizzell and Ruahama Weaver.  It also doesn't show a duplicate of Nathan and Ruahama in card view.  I can't remember if it did prior to the layout change or not.  Regardless they would need to handle pedigree collapse the same over all views of the new display.  I'm guessing it would do the same on a repeat grandfather who had children by multiple wives.  

You can change back to the classic view (at least for now) by selecting Thrulines card view option from the menu on the left. .

When you evaluate the connections a side bar pops out much like it does when you are running comparisons in Hints

Vertical View of match's line of descent from
Polly Jakes sister of my 3rd Gr Grandfather, James Jakes(1788-1847)
of Franklin Co., Virginia

View I see when I click on Samuel Bird, son of Polly Jakes

The image above shows the timeline view

Just below that is the option to toggle Sources which for this one included: 

Records (7) 

Ancestry Member Trees (3)

Photos (1)

Family (6)

What I Like:

• Multiple View Options: Flexibility to switch depending on research needs.
• At-a-Glance Overview: A larger, cleaner overview across several generations.
• Dynamic Zoom: Seamless zooming in and out on horizontal or vertical views.
• Better Layout of Details in the Evaluating Process

What I’d Like to See Next:

• An "Evaluate & Dismiss" Feature: The ability to mark a proposed connection as invalid and remove it from view.
• Parent Side Filtering: Automatically suppressing match suggestions that contradict verified Parent Side assignments. While this wouldn't eliminate incorrect connections that happen to sit on the right parent's side through a different line, it's a great start and would clear up many common errors.

Do you have the new view yet?  What are your thoughts?  

A Fun Exercise Using Thrulines--Descendants of Jabel Putman

I am fortunate to have tested myself and three of my siblings in an effort to capture the DNA we inherited from our Dad.  He passed away before autosomal DNA (atDNA) testing was widely available. That gives us four of his children who have tested. Each child received 50% of their DNA from him, but not the same 50%. Because of this, sometimes all four of us will share a common match. Other times, only one of us will have the match.

The table below shows the number of tested Jabel Putman descendants that Thrulines believes are connected through each child's branch, broken out across the four siblings. This was primarily a visual exercise in identifying potential connections through Jabel's children,  so I did not double-check every match. If you use this exercise to infer a connection through a Thruline ancestor, you will need to verify each connection independently. The table is a great illustration of the randomness of DNA inheritance. 

Jabel Putman(1772-1835) is my 3rd Great-Grandfather.  He was from Union County, South Carolina, before moving to Bedford County, Tennessee.

Things to remember:  

• Thrulines uses users' trees, and those could likely have errors that other trees have copied.
• Thrulines is side-blind and doesn't verify that the match is on the side of the family that would make the suggested connection possible.
• Thrulines links smaller segment matches, which may be false positives / identical by state and not by descent.

To make sure the Thruline suggestion is even possible:  

• Look at the shared matches you have with the DNA match connected with the Thrulines. Do they share with other known matches of that line?
• Is the match even on the correct side to make the Thrulines possible?
• Green dotted lines between each connection from the Thruline Ancestor to the DNA match indicate that it is a hypothesis and should be checked as thoroughly as possible before being used to make any inferences. The match may be connected through an entirely different line.

Studying the Thrulines gives you a better understanding of the matches because it forces you to prove or disprove the possibility of the connection through that match or line, which helps your overall research. Just as you never really know who might have the most Putman or Cook DNA, you also don't know who might have sources or clues to help with your research. Locating cousins can be very helpful, especially if they are actively researching. Everyone who tests isn't. I feel like this table will also be a helpful reference when I want to work on Jabel's line.