Unexcpected DNA Finding

Last week, I watched the "Watch Geoff Live: DNA" webinar on Legacy Family TreeWebinars. While watching, I followed along by looking at my autosomal DNA results on Ancestry. I had my DNA tested last summer and had studied some of the results but hadn't found the help with 'brick walls' that I had hoped for.

One of the brick walls I've been working on is my James Crawford line. My James Crawford died in Preble County Ohio in 1854. He married Sally Duggins in Garrard County Kentucky in 1799. Sally had previously been married to Alexander Duggins by whom she had two sons: William and Henry. Henry Duggins was married to Jane Sellers, daughter of Nathan Sellers. Nathan A. Sellers was a neighbor of my James Crawford and died in Preble County Ohio in 1824. In early Preble County Ohio, there was a second James Crawford who owned land adjoining my James Crawford. This James Crawford married Martha Knight in 1793 in Lincoln County Kentucky. It is believed that two of the sisters of this James Crawford married brothers to Nathan A. Sellers. Thus, the two Crawfords, the Sellers and the Duggins were all neighbors in Preble County Ohio and loosely related by marriages. However, no proof has been found of any relationship between the two James Crawfords.

As I was watching and following along, I somehow searched my DNA results for the Sellers surname. SURPRISINGLY, I found matches on Nathan Sellers of Pennsylvania! I've found at least two others (enough for a circle of 3) but the generations don't quite match up for Ancestry to create a circle (yet). Thus, I've started trying to document a circle.

A known descendant of James and Martha has his DNA on Gedmatch (but not on Ancestry). When we have compared our two sets of DNA, it said there were 'No Shared DNA segments found'. Thus, we assumed we either weren't related or that it was far enough back that the DNA didn't help.

With my new SELLERS discovery, I decided to re-look at Gedmatch. This time, I had it show the chromosomes. We still didn't have matches on 'centromeres'. However, the chromosomes weren't all red. Thus, I had to study the GedMatch key.

According to the key, the more lime green, the more matches on base pairs (A-T and C-G). When I looked at the chromosome comparison with the descendant of the other James Crawford, I found that we had a lot of yellow and green on chromosomes 21 and 22.

 When I did the same comparison between the descendant of James and Martha Crawford and a descendant of Nathan A. Sellers, there was even more yellow and green on these two chromosomes.

Granted, all results still say 'No shared DNA segments'. However, this is comparing autosomal DNA and not yDNA. I'm also looking at possible relationships at least six generations back. This yellow and green and the fact that SELLERS turns up as a surname match on Ancestry is enough to cause me to re-look at the research I have already done and use the SELLERS research to learn more about my CRAWFORD lineage.