Having delved into autosomal testing and followed many leads I can safely say that I'm related to a lot of people...but not very closely.
Autosomal:
My DNA is a genetic quagmire of humanity here in the states. Having my dad tested gave me a narrower field to search (for both sides) but personal revelations like sharing an X match with my father have made me realize that I need to narrow it even further. The X match isn't surprising given that my parents historic families live in the same towns on the east coast and intermarry a few times, it's just frustrating because it makes things a bit messier from a data standpoint. It means that my mom and paternal grandmother are related, which I can see in my tree.
Beside that hurdle I have a huge problem teasing my "Thompson" family from my "Seelye" family. I can't disentangle my Dad's parents without some way to break it down further back in time. So where does this leave me?
With both my paternal grandparents dead, I could ask a sibling of my paternal grandmother to test...not looking forward to that. I could also as a cousin of my father on that side to try to pull away what it means to be a "Seelye", but that would only do part of the job. Siblings don't carry the same mix of DNA so they aren't exact matches. The Seelyes are pretty well known and also well looked after by other researchers so, although the results would be meaningful, it would be like adding water to a lake when the Thompson puddle is drying up.
So that leaves finding someone on the Thompson side of my family. There are many options but none very close. My grandfather had no known siblings so I can't go to one of them to even try to get close. It would have to be a cousin or a descendent of a cousin.
In a way this would be very helpful. There is a known relationship and a known last common ancestor (Albert Thompson and Ida Williamson) and whatever portion of DNA these extended cousins shared with my father would be a pretty good indicator of "Thompson/Williamson-ness".
Why pretty good? Well, the down side is that there would be a lot of DNA missing and the very real possibility that the Thompson cousins are also genetically related to the Seelye family (made even more likely in my mind given that my own maternal family proves to be related to them).
With unlimited resources I would do both as that would give more pieces to play with but currently I can't support that kind of testing. In a perfect world I would pull samples from the Indiana Thompsons as well as Williamson and Blades related people to get a really good cross section of genetic relatives that I could play off each other to narrow down possibilities. In effect, I would hire ringers.