Matches who only tested at 23andme

Testing at 23andme has proven worthwhile, identifying some matches who only tested there vs on Ancestry, MyHeritage or FamilyTreeDNA but known to connect with documented branches on my tree on those sites. There are some familiar names too, cousins who had also tested on the other sites.

23andme has some tools that I like - the timeline, the reconstructed ancestors tools and clustering are useful. I like that it did a solid job of calculating my maternal haplogroup and allows you to see matches by haplogroup. There are also some features I find annoying, or are not as intuitive as Ancestry and MyHeritage. Looking at DNA Relatives (matches) is not as easy as Ancestry or MyHeritage, and not immediately clear which side of my tree these people are on. I think Ancestry does a much better job at separating out my maternal and paternal sides. 23andme does have a cool DNA Family Tree feature, but I am finding it's mixed up my Mexican side of the tree from my Mom's Whitley/Matthews side (these are both on my maternal side of the tree, but the Mexican side is maternal-maternal, and Whitley/Matthews is maternal-paternal).

23andme also limits you to 5000 DNA matches, vs the 82,574 that I have on Ancestry. But if this number is quality and represents people who only tested here and not on another platform, 23andme is providing potential connections I didn't have on Ancestry or MyHeritage. There's more to do in looking at the tools and possibilities.

And the results are in

 

23andme Ancestry landing page.

In the hopes of finding some potential connections who only tested at 23andme, I sent off a kit at the beginning of the month. Two weeks later, the results are in and I am looking at comparisons between other tests at Ancestry, MyHeritage and FamilyTreeDNA. 23andme nailed my legacy mtDNA haplogroup, A2d2, while FTDNA's mtFull sequence provides a more refined MitoTree result. For the paternal haplogroup, my Big Y 700 result is definitely more granular than the higher level R-L2.

Before comparing the Ancestry composition results with other tests, it's worth looking at a timeline report from 23andme (below).

23andme. Ancestral timeline.

The report estimates a most recent ancestor from a specific population based on inherited DNA segments. They estimate I had a parent, grandparent or great-grandparent who was 100% English between 1880-1940. Knowing my tree as I do, I know Samuel Oyler was born in Kent, England in 1822. Samuel's son Charles John Oyler (my 2nd-great-grandfather) was half German through my 3rd-great-grandmother Marianna Haiss.

The next interesting entry on the timeline says I had a great-grandparent, 2nd-great-grandparent or 3rd-great-grandparent who was 100% Andalusian, Asturian or Castilian. Here is where some of what I know on my Mexican sides of the tree is useful but incomplete on several of the branches. At the same level, 23andme suggests I also had someone who was 100% Portuguese or Galician, born between 1820 and 1880. I think this was further back, knowing how mixed the people in my tree were about this time.

Between 1760 and 1850, I had most likely a 2nd to 5th-great-grandparent who was Irish. Yes! This would be my 3rd-great-grandmother Bridget, and John O'Brien. 23andme estimates I also had someone at this level who was 100% Scottish. Yes! This was my 5th-great-grandfather John McIntosh. 23andme similarly suggests I had someone at this level who was 100% Austrian or Southern German. Correct, this could come from Marianna Haiss, who was born in Hausen im Killertal in Southern Germany.

Further down in the timeline, there are some fascinating recommendations. 23andme suggests I had a 3rd to 7th-great-grandparent or greater who lived between 1700 and 1820 who was 100% Nigerian. Ancestry's latest origins update from October 2025 estimates this very likely enslaved person was from the Nigerian woodlands, so it's interesting 23andme also shows this trace connection. At this level it also suggests an ancestor who was 100% Southern Mesoamerican (possibly Mayan).

At the lowest level of the timeline, 23andme suggests between 1700-1790, I had an ancestor who was 100% Ashkenazi Jewish, one who was 100% Northern Andean (this is from the Chilean branch of the tree), someone who was 100% French, someone who was 100% Belgian, Rheinlander or Southern Dutch - yes I have this from the du Trieux branch.

Your ancestors aren't who you think they are

 

YouTube. David Reich interview. 21 Apr 2026.

There's a fascinating talk from Harvard geneticist David Reich that was recently uploaded to YouTube. The talk is 49 minutes, but it is broken into sections. Reich published a book in 2018 titled Who We Are and How We Got Here, and the YouTube interview is based on his team's research and the themes of the book.

Nature article on DNA and Indigenous Americans

A new article published 22 April 2026 in Nature describes the Evolutionary History and Unique Genetic Diversity of Indigenous Americans. This is the largest genomic dataset of Indigenous Americans to date, helping researchers trace the threads of early migration across the Americas.

It will be interesting to see if this dataset improves future versions of ancestral regions reports from Ancestry, MyHeritage and others. These companies use reference populations in their models, and as the data improves, this can provide more accuracy to these ancestral regions reports.

Through the brick walls

 

Photo by Patrick Jones. Farnham, UK. 23 Apr 2026.