I was excited this week to see a new close relative, Sal, in my DNA matches on FTDNA. Sal is my third closest match to date, and we share 149 cM of DNA, suggesting that, on average, we will be 3rd cousins once removed or more closely related.
Sal has a wonderful ancestral tree on FTDNA, complete back at least 5 generations, so it should have been straightforward to determine how we are connected, or at least to identify a shared geographical location. As I searched Sal’s tree, I noticed that all of his ancestry is from the Italian island of Sicily. Perhaps we share some of my 3% Italian ethnicity that MyHeritage (but no other testing company) has identified?
The next step was to look at exactly which segments of DNA we share, and that is when it all fell apart….
When I opened the chromosome browser, it showed that Sal and I only share one 8.8 cM segment on chromosome 3.
What happened to the other 140 cM that we share? The FTDNA chromosome browser offers a choice of 4 options for the minimum segment size which will be displayed (1, 5, 7 and 10 cM) with the default being 5 cM. All but 1 of the 50 segments that Sal and I share are less than 5 cM in length, rendering them insignificant in determining recent connection. Sal has gone from being a likely 2nd cousin once removed to possibly an 8th cousin or less – a connection that we are unlikely to be able to identify through paper research.
I have not come across this situation before with chromosomes 1 thru 22, although I have seen a similar issue on 23andMe with chromosome 23 (X). 23andMe includes any cM shared on the x-chromosome in the total amount of DNA which they identify as being shared by two DNA testers, despite the inheritance pattern of x-chromosome DNA being totally different from the other 22 chromosomes, and it varies between males and females. My wife has a group of DNA matches high on her 23andMe match list, which would not feature as prominently at the other testing companies. e.g. She and Tina share a total of 88 cM, but 42 of these are on the X-chromosome; she and Lisa share a total of 84 cM, but 62 of these are on the X-chromosome.
The chart below is an over-simplification, but shows generally how the various testing companies handle small segments and X-chromosome DNA.
So, not all shared centiMorgans are equal and, especially on 23andMe and FTDNA, may require more analysis to determine the closeness of the relationship with a DNA match.