Reconstructing My Great-Grandparents (As The Creeper Tour Rolls On)

Kevin and Rolf exhibiting Borland Genetics in Sweden

As many readers may be aware, Borland Genetics has been promoting uploads to the site throughout 2022 via a hybrid in-person/virtual string of appearances at industry tradeshows, genealogy society events, genealogy vlogs, etc., all under the penumbra of the "2022 Creeper Tour."  So far, the tour has passed through Baltimore, Cincinnati, Sacramento, Ottawa, Indianapolis, Portland (Oregon), Skövde (Sweden), and Burbank, and upcoming dates are scheduled for Minneapolis (virtual), Sydney (Australia), Salt Lake City, Richmond, Sandusky, Milwaukee, and finally back to Baltimore (as the Tour extends through 2023).

Some of the dates thus far have been expo hall exhibits with opportunity to meet me in-person and discuss Borland Genetics reconstruction techniques, while others took the form of virtual talks where I presented topics including DNA inheritance, DNA reconstruction, interpreting one's DNA results, the future of DNA tools, and using the Borland Genetics tools and database.

Now, as the Creeper tour rolls into the autumn, a new series of talks is in the making, where I will discuss specific workflows I used to reconstruct DNA profiles of my own great-grandparents (and beyond).  For example, in October, I will be presenting "Reconstructing Max Freudenberg" for the Minnesota Genealogical Society, and I will later present "Reconstructing Clara Dzyban" for the Czechoslovakian Genealogical Society International in Milwaukee.

So, part of the rationale for writing this post is of course to promote the upcoming events (the details of which will be posted in the Borland Genetics Users Group on Facebook as each event draws nearer).  Also, it's a good opportunity for me to post a snapshot of how my personal reconstruction projects are coming along.  I made the following chart to show the results I've gotten so far:

My ancestors are represented by ahnentafel numbers, and the results are indicated in terms of percent coverage.  As you see, I was able to reconstruct 71% of my father Steven Borland using as input the factory kits of two of his descendants (my brother and I).  In addition, for this reconstruction I also incorporated my mother's DNA kit which was used to phase both my brother and I across our entire genomes, facilitating the reconstructing process.

At the grandparent level, I was only able to reconstruct 35% of my paternal grandfather John Borland so far, not coincidentally very close to half of the 71% coverage of my father's reconstruction, because I'm only relying on the same 2 descendants, i.e. my brother and I.  However, my other three grandparents are in the 52-54% range because I was able to incorporate data from additional descendants.  I hope to increase John Borland's reconstruction coverage by recruiting one or more of his descendants to test and upload in the future.

At the great-grandparent generation, the reconstruction coverages obtained vary greatly.  Part of the explanation is the variation in number of tested descendants (where a 2-descendant reconstruction of Elizabeth Forbes yields a mere 3%, whereas a 9-descendant reconstruction of Arthur Freudenberg yields 28% of his genome).  However, there are other factors at play affecting the outcomes.  At this range, we also rely heavily on cousins in the database to assign parental phase of segments as we're drilling back generation by generation.  So, a major factor in whether the output coverage is high or low is the number of discernable cousins that are in the Borland Genetics database, or who have their DNA on other sites that report segment start/stop points where they match with the descendants.  It is for this reason that I strongly encourage people to upload to Borland Genetics for free, even if they don't intend to use the advanced tools on the site.  Cousin data is extremely important to the reconstruction process, and so the size of the database directly affects users' ability to do deeper reconstructions (as a larger database means more cousin matches in the system).  So, going around the world exposing people to Borland Genetics, encouraging uploads, and educating people about DNA reconstruction, is an important part of my job at Borland Genetics, just as important as developing new tools and features.  Resources like the Borland Genetics database grow more valuable to the genetic genealogy community as more of the community participates.

There are some other factors unique to specific ancestors that make them easier or harder to reconstruct.  For example, Eva Douse was an immigrant ancestor from Lithuania, and testing is not especially popular in Lithuania, so there are not likely to ever be many of her cousins in the database upon which I can rely for assigning phase.  On the opposite end of the spectrum is Arthur Freudenberg, who was half Jewish and has many, many cousins in the Borland Genetics database due to endogamy.  And then there's the couple Weldon Borland and Elizabeth Forbes that not only have few tested descendants, but also happen to be related to one another and share many of the same Quaker ancestors, so distinguishing the cousins that pertain to one vs. the other is very difficult.

Anyway, hopefully as you're reading this, you are considering uploading your DNA kits to Borland Genetics and trying out some of the reconstruction tools.  There are a lot of instructional videos available from the Tools tab of the site, and as I've mentioned, I'm going around the country (and the world) giving talks on how to get more out of the site and having individual discussions with site users.  I hope to see more of you as the Creeper Tour rolls into autumn and then 2023!