Category: Genetic Genealogy

Incorporating DNA into your family history research

Posted on

2024 – A Good Year to Be a Genetic Genealogist

Every year brings new tools, websites, and approaches for genealogists to try. 2024 was exceptional! Here are four innovations that transformed my work, ranked by their impact:

1. BanyanDNA

Leah Larkin (the DNA Geek) and Margaret Press (of the DNA Doe Project) teamed up with developers and statisticians to create BanyanDNA, a tool many of us didn’t know we needed. This web-based app lets you draw complex family trees and check if the shared cM amounts match your diagram. And you don’t have to limit yourself to one DNA tester. Any pairwise comparison of DNA between any two people on the tree can be included.

BanyanDNA has become essential for me. I can run 1,000–10,000 calculations to simulate recombination randomness, validating my tree or finding the best placement for a mystery person. It even handles intricate relationships, like cousin marriages or double connections.

For example, I mapped a cousin marriage between the children of two of my 3x-great-grandparent couples: Holloway Key and his wife Catherine, and Thomas Adam Walker and his wife Julina Allen who lived in Benton County, Tennessee. In the image above I used red lines to show the descendants of the Key couple and blue lines to show the descendants of the Walker-Allen couple. The pink and purple lines represent where the two Key brothers married two Walker sisters. My mother, Gladys, and two of her siblings are DNA testers and I’ve shown three of their matches who are related to them through both couples. This kind of situation can throw off the expected amount of shared DNA, but BanyanDNA can handle it. Below is a validation run that shows me that there is one match that falls outside the expected range of shared DNA. Often when I see that result, it means I have drawn the tree wrong or mistyped the amount of shared DNA.

Summary:

I use BanyanDNA to:

  • Diagram complex trees
  • Record shared DNA between large numbers of DNA testers
  • Validate my work using statistical simulations
  • Place unknown matches on the tree.

What you need:

  • A good understanding of relationships and family trees
  • Shared DNA amounts (see Enhanced Shared Matching below)
  • A willingness to learn how to use BanyanDNA. There are recorded tutorials here, a BanyanDNA User Facebook group, and office hours for free (publicized on the Facebook group). Leah Larkin offers a webinar tutorial for a small fee.
  • Basic statistical skills to interpret the calculations.
  • Patience! Large trees with many matches take time!

2. Enhanced Shared Matching at Ancestry

MyHeritage has long offered excellent tools for genetic genealogists, like downloadable DNA segments and AutoClusters. This year, Ancestry followed MyHeritage’s lead and introduced Enhanced Shared Matching, which shows the shared DNA between you, your matches, and their mutual connections.

The image above is a privatized look at Enhanced Shared Matching at Ancestry. My DNA match is CP and I know how I am related to her. With Enhanced Shared Matching (available with an additional subscription), I can see not only how much DNA I share with our shared matches, I can now see how much DNA CP shares with them. I sorted this list by how much they share with CP which helps me discover more about her family. For example, the first match, a, shares 603 cMs with CP but only 51 cMs with me. 603 is in the range of a first cousin to CP. Match a has a 9 person tree with only two visible people on the tree whose names are not found in my family tree. But knowing they are possibly a first cousin to CP means I can likely figure out their exact relationship to me.

All of the Enhanced Shared Matching data can be entered into BanyanDNA so I can validate this branch of my family tree.

Summary:

I use Enhanced Shared Matching to:

  • Understand how my matches are related to each other.
  • Place matches without full trees on my tree.
  • Supply data for BanyanDNA calculations.

What you need:

  • A ProTools subscription (currently $10 USD/month)
  • Knowledge of shared DNA amounts
  • Tree-building skills
  • Ability to research DNA matches with incomplete or no trees.

3. Full text search at FamilySearch

FamilySearch, the world’s largest free genealogy resource expanded its capabilities in 2024. Its new Full Text search of historical records (available at FamilySearch Labs) uses AI-driven optical and handwriting recognition to unlock thousands of previously unindexed records, such as land deeds and probate files. For those working on reparative genealogy, the Freedmen’s Bureau Records are part of the collection that can be accessed.

From the FamilySearch Labs home page, click on “Go To Experiment” under the Full Text option. This will take you to a search page.

Use the keyword search to enter your names – that provides you with more opportunity to find your ancestors. I entered Calvin Whitney in Keywords, and Maine from 1800-1870 in the year range to discover the record below:

Calvin Whitney is my 4x great-grandfather. I don’t know a lot about him other than he died in October of 1832 in Thomaston, Lincoln County, Maine. His son, William E. Whitney, left Thomaston in October of 1849 for the California gold rush and later built four lime kilns at the foot of Mt. Diablo in Contra Costa County, California. This deed describes the property that Calvin Whitney purchased from David Watson in November of 1827 and that Whitney had “the liberty of digging & carrying away the lime rock from said land.”

Summary:

I use the FamilySearch Labs Full Text Search to:

  • Find digitized records that are not indexed

What you need:

  • A search strategy using keywords, places, and date ranges.
  • Patience-more records are added regularly

This brings us to the final aspect of my genealogy work that changed in 2024: my use of AI.

4. Artificial Intelligence (AI) and genealogy

AI is revolutionizing genealogy. In late 2023, I joined Steve Little when he first offered the “AI for Genealogists” course through the National Genealogy Society. In the course I learned how to do the basics with AI: summarize, extract, generate, and translate. Much of my use of AI was confined to making fun images in my brand colour scheme for presentations.

Image created by DALL-E and me.

The potential for AI became clear during a reparative genealogy project. I was provided with a Civil War diary of the ancestor of my client. This project documents the people enslaved by the client’s family on Gwynn’s Island, Virginia. I used ChatGPT to transcribe the diary and to create lists of the people and places mentioned in the diary. AI saved me hours, though I still had to check its work. I asked ChatGPT to write a commentary on the contents of the diary based on historical context. It was fascinating! Here’s the conclusion:

Conclusion

The diary of Mary T. Hunley is a compelling window into the experiences of Confederate civilians during the Civil War. While shaped by the author’s biases and Confederate loyalty, it highlights the war’s disruptive nature, the agency of enslaved people, and the fragility of Southern society. It stands as both a testament to personal endurance and a reflection of the deeply ingrained injustices of the antebellum South.

I have a long way to go before I do some of the amazing things genealogists are doing with AI. There’s a great Facebook Group, Genealogy and Artificial Intelligence (AI), Steve Little blogs and speaks regularly, and there’s a podcast, The Family History AI Show, to help you keep current. Nicole Dyer and Diana Elder at Family Locket also teach, blog, and podcast about using AI in their genealogy work.

Summary

I use AI tools to:

  • Transcribe handwritten documents
  • Summarize lengthy articles or records.
  • Create tables and lists from documents.
  • Create images.
  • Understand historical context.
  • Edit and refine my writing (including for this blog entry!)

What you need:

  • Awareness of privacy and copyright concerns.
  • Knowledge of Large Language Models (LLMs).
  • Practice, practice, practice.
  • Careful review of AI outputs

Final Thoughts

Genealogy is ever-evolving, and 2024 raised the bar with transformative tools like BanyanDNA, Enhanced Shared Matching, Full Text Search, and AI. Each one has enriched my research and streamlined my workflows. What’s next for genealogists? I can’t wait to find out!

Posted on

Let’s contribute to science!

I’ve always been a fan of what used to be called “Citizen Science” and has recently been re-named “Community Science.” I contribute birding data to ebird. I was a volunteer atlasser for the BC Breeding Bird Survey in its early days and I continue to support my husband on the annual survey. We start at the crack of dawn and he counts birds while I drive and keep the tally.

I’m excited to have the opportunity to support genetic genealogy. Blaine Bettinger, the genetic genealogist who is behind the Shared cM Project on DNA Painter, has opened submissions for the next update. The goal is over 100,000 submissions. He hopes to release the update in early 2023. Now’s the time to contribute! There are two ways to do it.

  • Through this spreadsheet link which is a great option of you have a lot of data.
  • Or one entry at a time using this Google Form.

If you are unfamiliar with the Shared cM Project, it is the go-to place to check genetics vs. genealogy. That’s the step in the analysis when you see if you and your DNA match share the amount of cM that will help you confirm or figure out a relationship. You can get these predictions from each of the testing companies. That’s the notation by your match that says “2nd-3rd Cousin.” Each company uses their own data and information, which could be based on faulty trees. What’s different about the Shared cM data is that it is crowd-sourced. Genealogists who have confirmed the relationships through documentary research provide the data. That’s us! This will be my third time contributing data.

Let’s get to work!

Posted on

Ancestry’s Ethnicity Inheritance Chromosome Painter and Jonny Perl’s new tool

The latest genetic genealogy features and tools bring excitement and potential for new insights. The combination of Ancestry’s Ethnicity Inheritance Chromosome Painter and Jonny Perl’s ACPS (Ancestry Chromosome Painter Segments) tool allows comparison of Ancestry data with information from other testing companies. Jonny’s new tool uses the information from Ancestry to grab <<gasp>> segment data <<gasp>>. It’s not the segment data you’ve been hoping for (segments shared with matches) but it’s fun and potentially useful depending on your situation.

Several months ago, Ancestry released ethnicity results with “Side View” technology. Side View allows you to see which ethnicity you received from each of your two parents. Unless you’ve had a parent tested or recognize which parent is which, the two parents are simply labelled “Parent 1” and “Parent 2.” You can label them yourself.

Here’s my mother’s ethnicity split into the contributions from her two parents.

Ancestry Ethnicity Inheritance Overview Showing ethnicity for each parent of my mother

Ancestry is now rolling out the beta feature of a chromosome painter for ethnicity, catching up with 23andMe and FamilyTreeDNA who already offer this feature.

My mother’s results on Ancestry look like this:

Ancestry’s Ethnicity Inheritance Chromosome Painter for my mother

Each numbered pair represents my mother’s chromosomes, 1-22 (chromosome 23 is the chromosome that determines biological gender and Ancestry does not show the results of chromosome 23.) One chromosome of the pair is from my mother’s dad and one chromosome from her mother. The different colours represent different ethnicities, which refer to geographic locations. As you can see, my mother’s ancestry is primarily European, and largely from England and Northwestern Europe. As of this moment, I don’t know which of the two (upper or lower chromosome) is my maternal grandfather and which is my maternal grandmother. My mom has three segments labelled “Indigenous Americas-North” and one labelled “Southern India.”

My maternal uncle’s Ancestry Ethnicity Inheritance Chromosome Painter results look like this:

Ancestry’s Ethnicity Inheritance Chromosome Painter for my maternal uncle

Notice the colours in the key are not stable, which is a shortcoming of the Ancestry Chromosome Painter and their ethnicity results in general. It would be useful for Norway to be the same colour no matter whose DNA test I am looking at. Norway is blue for my mom and purple for my uncle. And then there are the two shades of purple that are similar – Norway and Indigenous Americas for my uncle. And all the shades of green, aqua, blue, darker aqua… more variety would be helpful. I’m grateful for the information and I can edit the colours when I move the segment data to DNA Painter using Jonny’s new tool. My uncle has four segments labelled Indigenous Americas. They are on chromosomes 3, 4, 6, and 12.

Here is my maternal aunt:

Ancestry’s Ethnicity Inheritance Chromosome Painter for my maternal aunt

My aunt has a new category, Germanic Europe, and two Indigenous America’s segments. When comparing the three images, it’s important to remember that Ancestry randomly assigned Parent 1 (upper chromosome) and Parent 2 (lower chromosome). My mother’s seems to be reversed from her siblings since the “Indigenous Americas” segment on chromosome 3 is Parent 1 for my mom and Parent 2 for my uncle and aunt.

Jonny Perl’s new tool allows me to copy and paste the data that creates the Ancestry image and make a file I can manipulate at DNA Painter. (To learn how to use DNA Painter, there are several good webinars with demonstrations and resources here.) Jonny used known information about DNA to approximate the start and stop points from the Ancestry chromosome images to make this work.

Ethnicity results can be compared across the companies. Remember my uncle who had several chromosomes with Indigenous Americas on Ancestry? Here I’ve uploaded his FamilyTree DNA ethnicities on the same DNA Painter diagram as his Ancestry ethnicity. I’m showing the detail for chromosomes 3 and 4.

FamilyTree DNA and Ancestry Ethnicity results compared for my maternal uncle

FamilyTree DNA results for both maternal and paternal chromosomes are mapped to “Shared or Both” because I don’t know which chromosome is maternal and which is paternal. Ancestry DNA chromosomes are artificially labeled maternal and paternal. FamilyTree DNA uses fewer ethnicity labels than Ancestry. Nearly all of my uncle’s DNA is labelled “Western Europe” and shows up as a darker pink bar. Ancestry separates “England & Northwestern Europe” shown in green and Scotland in aqua. FTDNA’s “Americas” segments in green are not as extensive as the approximated “Indigenous Americas-North” from Ancestry via Jonny Perl’s new tool. This may change if Ancestry ever allows a real download of raw ethnicity data. The Indigenous Ancestry is a small percentage of my uncle’s DNA but seeing it detected by both companies increases my confidence in the segments.

The segment on my mother’s chromosome 5 that is attributed to Southern India caught my attention. I’ve painted matches from GEDmatch, MyHeritage, and FamilyTreeDNA to DNA Painter. (My mom is not on 23andMe, which is the other testing site with segment data.). I was curious to see if there were any matches corresponding to that unusual segment.

Chromosome 5 ethnicity and matches for my mother

Remember that the Ancestry ethnicity segments which are imported as maternal or paternal could be the other way around. In this case, there is an unknown maternal match for SC (in bright pink) that does seem quite similar to the Southern India segment. I will keep an eye on this, it could be helpful in the future.

I’m encouraged by the new developments and hope they unlock mysteries for you.

Posted on

AncestryDNA Communities and the Benefits of Testing the Oldest Generation

I look forward to learning about the new tools and updates that come from DNA testing companies. Recently Ancestry.com updated their genetic communities. Genetic communities demonstrate the link between genetic genealogy and traditional documentary genealogy research. Ancestry uses the family trees of customers and the power of their massive autosomal DNA database (over 21 million testers) to place groups of ancestors in a place and time. The time frame (starting about 300 years ago) matches the time period when documentary evidence might be found. DNA communities provide hints for further research.

Autosomal DNA inheritance is random and siblings (unless they are identical twins) will have different results. My brother and I each received half of our atDNA from our dad and half from our mom, but we didn’t get the same half.

My results aren’t very impressive:

My DNA Communities from Ancestry.com

I have one community. Clicking through the information tells me which DNA matches and which ancestors fit in this group.

Here are the results for my brother:

My Brother’s DNA Communities from Ancestry.com

The DNA Community for my brother is different than the one I have. One of the same 4x-great grandfathers on my maternal side, Holloway Key (1777-1855), is in both groups. This makes sense since the geographic area of Tennessee where he settled appears in both. For my paternal side, there are hints I can pursue in the Lower Midwest & Virginia Settlers group for the House family that I am curious about.

I tested my mother before she passed away, and her brother (my maternal uncle) also agreed to test. That puts me one generation closer to my maternal ancestors.

Here are my mother’s results:

My Mother’s DNA Communities on Ancestry

Not only does my mother have the DNA community my brother had, Southern Midwestern Settlers (his is orange, hers is green), she has four others.

My uncle provided an additional community, Early Upper South Settlers:

My Maternal Uncle’s DNA Communities on Ancestry

The results differ a little, and that’s because of the different autosomal DNA they inherited from their parents (my maternal grandparents). The number of new communities demonstrates the rich genetic information older generations hold.

Here’s a comparison of the results for the four of us:

RegionMotherUncleMeBrother
Arkansas, Oklahoma & Texas Settlers (1700-1975)√√  
Early Alabama, Mississippi, Louisiana & East Texas (1700-1800)√   
Early Upper South Settlers (1700-1975) √  
Lower Midwest & Virginia Settlers (1700-1950)  √ 
North Upland South Settlers (1775-1975 Northwest Tennessee & Kentucky Jackson Purchase Settlers√√  
South Central Appalachia Settlers (1700-1950) North Carolina High Country Settlers√√  
Southern Midwestern Settlers (1700-1975)√√ √
Comparison of DNA Communities for My Mom, My Maternal Uncle, Myself and My Brother at Ancestry

The text describing the community, animated maps showing migration patterns, and hints from Ancestry provide clues that I can research. We don’t know anything about my maternal 2x great-grandmother, Mattie (Childres) Fisher Pike Adams outside of 3 marriages and 1 census. Ancestry placed this branch in the “Arkansas, Oklahoma and Texas Settlers” with origins in Northwestern Europe and the British Isles, possibly migrating to the Chesapeake Bay area and moving across Tennessee and Kentucky. This is consistent with the Childres matches I am researching.

If I did not have my mother and my uncle, I would I would be missing five genetic communities. Perhaps in the next update, the DNA communities will show up for my brother and me. I welcome the new clues in my search for Mattie’s family.

Posted on

That new project feeling

Every new project has a honeymoon period. There’s the anticipation of what might be discovered and the pleasure of setting up the project so that the research proceeds smoothly. Here’s a glimpse into my project set-up process.

Research Log

A little over a year ago I started using Airtable as my research log. I have enough understanding of programming to be dangerous. I even took a class in Fortran in college and I learned how to use Microsoft Access when I worked in a health services research setting. I love relational databases even more than I love spreadsheets. Airtable is a “spreadsheet-database hybrid with features of a database but applied to a spreadsheet” according to Wikipedia. Nicole Dyer of Research Like a Pro (RLP) with FamilyLocket has shared terrific bases (templates) in the Airtable Universe. I developed my own adaptations to Nicole’s RLP with DNA Multiple Testers Base.

Below is a screenshot of the documentary research log I use based on Nicole’s template. I’ve grouped this log by surname (4th column), then sorted it by date of event (3rd column) then locality (5th column). Dates are written in YYYY-MM-DD format to sort the records (rows) by a field (column). Place names begin with the two-digit state abbreviation, followed by county, and town. This arrangement also supports sorting without creating multiple fields to sort. And if you are looking at column 8, yes, I do create the citations when I first look at the document. I have a citation template (also in Airtable).

Documentary Research Log in AirTable

Digital Filing

Every project has a folder on my drive which is backed up to Dropbox. The folder organization varies by project. The snapshot below is for my mysterious 2x-great grandmother, Mattie Childress. Each research question has a numbered file. “00-Mattie death date” and “01-Parents for Mattie” are examples. Nested within the folder are sub-folders for phases of research. “Archive” holds prior drafts and will be discarded at the end of a project. “Analysis” contains tables for census or timeline analysis. “Images” has screenshots of diagrams.

Digital Filing System

I’ve used Google Docs and Sheets and struggle to use the filing system in Google Drive. Many people use them to good results. Using Google docs, sheets and drive allows sharing through links. I get around this by sharing Dropbox folders for large files.

Naming conventions

File names for individual items (documents, images, spreadsheets) have evolved. There are two styles I use consistently: Name of document first or date first. When I am writing a report, the name of the document is written with underscores, then the date, always in YYYY-MM-DD format.

Childres_father_2022-01-31.docx

For sources (scans of originals, downloaded census, birth, marriage, death, etc.) I lead with the date, followed by the last name in caps, then location and type of record. I use dashes between elements of the date in YYYY-MM-DD format. This creates an instant timeline in my folders. Locations are organized from large geographic region to small.

1879-05-08_CHILDRESS_Bill_TN_Tipton_indictment_riot.jpeg

Report template

I set up the project report using a template and “write as I go.” Writing the report while I’m working supports the analysis process and I’m not faced with a blank page at the end. When working on client projects, writing-as-I-go also helps me stay within hourly limits for a project. The template, like naming conventions, has evolved over time. I use a Microsoft Word .dotx file which includes sample paragraphs I often need (such as explaining the usages of different types of DNA tests or how autosomal DNA is used to predict relationships).

Report template

I started a new project two weeks ago and set it up as described here. I can spend a few minutes or several hours researching knowing that I can focus on the research and analysis. And I’m enjoying the honeymoon!