I had a unusual scenario where I had two surname clusters that probably share a common ancestor with the surname of Casey. I knew that these two clusters would probably never be connected via traditional research as the connections are just too far back in time (600 to 800 years ago). However, I just knew that there had to be something to be gained by attempting to connecting these clusters that would add to analysis of each surname cluster.
My surname cluster was the South Carolina Casey cluster which has around 20 very closely related submissions - most at 67 markers. Around half of the submissions had 460=12 and the other half had 460=13 - both of these marker values are very rare for the R1b haplogroup and established an early branch which is very useful for genealogical research. Plus one submission had 460=11 which is not that rare. The second Casey cluster was labeled Munster, Ireland cluster (dominated by submissions with origins to that area) and this cluster had the more common 460=11. It seemed obvious that 460=12 was probably the MRCA (older marker value) for the South Carolina cluster. By knowing that the other cluster was 460=11, I became more comfortable that the MRCA was 460=12 for the South Carolina Casey cluster. But I wanted some genetic source documentation that would support this speculative conclusion. I wanted to really know if these two surname clusters really shared a common ancestor that used the surname Casey. If the two clusters were related, I also wondered how (and if) each cluster would assist my genealogical research if we knew how these clusters were connected.
With the above in mind, I decided to see how (and if) deep ancestry researchers analyze these scenarios. Deep ancestry researchers are focused on creating new broad branches of the Y-Haplotree and most care very little about genealogy which was a major culture shock to me. Four or five years ago, my oldest SNP was over 4,000 years old and just did see how it would help my genealogical research. Around one year ago, L226 was discovered for my cluster and it is only around 1,000 to 1,500 years old - getting much closer to the genealogical time frame - this seemed close enough to the genealogical time frame.
Deep ancestry researchers have created huge spreadsheets of submissions that have been tested for their haplogroup and are believed to be part of the haplogroup. Many haplogroup admins also create many DNA fingerprints (mutations from the MRCA haplogroup) in an attempt to discover new haplotree branches. They are fixated on these DNA fingerprints and tend to really ignore any mutational difference as it varies so much is not reliable. They even search Y-Search using DNA fingerprints to build their databases and ignore mutational difference.
Here is what I found. Armed with DNA fingerprint of each of my surname clusters, I quickly found a third related cluster with the surname of Kersey. This lone submission had many of the very unique marker values of the South Carolina cluster - huge major genealogical breakthrough. Armed with the MRCA of the L226 haplogroup (never really analyzed before), I quickly created DNA descendancy chart that shows how these clusters must be connected - if they are indeed related. To prove that these three clusters were indeed related, I analyzed and reorganized the L226 spreadsheet by surname clusters that were related and came up with 14 well defined clusters - around half of all L226 submissions. I verified that all three Casey / Kersey clusters have unique branches off of L226 from all the other surname clusters - confirming that the mutations for the Casey clusters are unique from all other surname clusters.
I also discovered that my South Carolina DNA fingerprint (the mutations from the L226 MRCA to the MRCA of my South Carolina cluster) was a much better way to screen possible NPEs. Matching DNA fingerprints (sharing unique mutations) is much more important than mutational difference between submissions. I also discovered "private" SNPs and how deep ancestry researchers find new SNPs. Private SNPs are ignored by deep ancestry researchers as they only want to define new branches on the haplotree and these SNPs are only 100 to 800 years old. The ignore these SNPs since they only create branches within surname clusters. These private SNPs are ideal for genealogical research.
The deep ancestry researchers place special "Walk The Y" tests from Family Tree DNA. These are partial Y-chro scans that include thousands of SNPs. Half of the people that order this test find new SNPs !!!! I plan on ordering a WTY test for my L226 haplogroup in hopes of finding either yet a more recent haplogroup or a private SNP within my South Carolina cluster. Sometimes multiple private SNPs are found and most genealogists are not even aware of these extremely important private SNPs.