T-SNP - My Fix

Terminal SNPs - arrghhh! 

I actually like the idea of working with Terminal SNPs (called T-SNPs for the rest of this blog).  But - without a companion "long form" Haplogroup to give them meaning, they are confusing and difficult to use.

After puzzilng over this for a while, I came up with my solution - one that will be used on Project Results tables at WorldFamilies that are prepared with our Results Tool. 

My solution is to provide two columns:

  • The T-SNP reported by FTDNA
  • The companion "long form" Haplogroup identified for that T-SNP by ISOGG

With these two columns side-by-side, a project member or researcher can see immediately which T-SNP is the most recent.  (the true T-SNP for the Lineage until another, more recent SNP is found)

There are a few imperfections:

  1. Sometimes, the T-SNP reported by FTDNA is not on the ISOGG tree.  (The Tool reports "Not Found" in the ISOGG Long Form column)
  2. Sometimes, the implied tree structure at FTDNA doesn't align with ISOGG (The Tool ignores this since FTDNA does not report a "Long Form" Haplogroup)
  3. Sometimes, FTDNA and ISOGG use a different SNP to define the same branch (No problem as long as both SNPs are reported at ISOGG - then the correct long form branch is reported)
  4. Sometimes ISOGG has developed their tree to show more recent branches based on SNPs that a man has also tested positive (The Tool doesn't know of the most relevant positive SNP - because FTDNA did not report it as their T-SNP)
  5. FTDNA is still tweaking their T-SNP estimates and tree structure - so it may be different from the WorldFamilies Results Table  (The Tool used the T-SNP reported the day the Results were updated)
  6. ISOGG is constanting updating their tree - so it may be different from the WorldFamilies Results Table (The Tool used the companion long form to the T-SNP reported the day the Results were updated)

T-SNPs also have another problem - none are the true Terminal SNP

  • The tested SNP is not the man's "Terminal SNP" - it's simply the most recent SNP that has been identified by testing AND given a place on the FTDNA tree.  All of us expect to find even more recent SNPs as more of the genome is tested.  (Someday, we will all be able to know our true T-SNP - but it actually won't be that useful, as only our immediate family will have it.)
  • The FTDNA estimated T-SNPs are somewhere in the man's long form Haplogroup - but almost always not "terminal" - when the most recent SNP for the man's genetic group is considered

Our New Format:  I am working my way through updating the project results pages that use our WorldFamilies display and are prepared with our Results Tool.  If you want to see this new format - here are examples (best viewed in Internet Explorer)

Terry

Whitman Family

Does the Whitman Family website still have an administrator? Looks like the last action on the site was several years ago. I just joined in January, but have yet to see any of my records uploaded.

New Format

I like it!  Thank you.

T-SNP Fix

Terry,
I suggest a compromise using the first letters of the Haplo assignment plus the "short hand"/T-SNP such as:

Ria-M417, or R-417
Rib-M269
I-P37
etc.

This takes only one space but identifies the original haplo groups which have been used very successfully to group related individuals plus the latest SNP for eacch person.

The short hand/ SNPs still leave me with some problems. I am Ria-417 along with 2 of my sons and 1 g-nephew, but my brother and 2 other nephews are Ria-M198. All 3 nephews are 2 sons and a g-son of another brother, who is deceased and untested.

I guess i still have a lot to learn.