mtDNA HVR2 146 marker on yFull/mTree

[DJ Brooks]

Can anyone with mtDNA knowledge make sense of this?

https://www.yfull.com/mtree/L3e2a1b1a1b/

How can there be a T146C (T to C) mutation, when the rCRS reference value of 146 is C by default. If it's currently C as default according to the rCRS wouldn't it being C not even register as a unique mutation? How is a T to C mutation identified as such, when it's doesn't have a default position of T to begin with?

[nooni]

It shows T146C! rather than T146C, which you're right about not existing. The exclamation mark is what's used to indicate a back mutation. C was the default value, it mutated to T, and then back to C. You'll sometimes see more than one exclamation symbol, so something like C146T!! would mean that the marker mutated three times on its way to the final value.

[DJ Brooks]

It shows T146C! rather than T146C, which you're right about not existing. The exclamation mark is what's used to indicate a back mutation. C was the default value, it mutated to T, and then back to C. You'll sometimes see more than one exclamation symbol, so something like C146T!! would mean that the marker mutated three times on its way to the final value.

How do they know it back mutated? C146T doesn't appear to be a defining mutation for the parent clade. My haplotype is the parent clade of the one linked and my value is 146C.

[nooni]

How do they know it back mutated? C146T doesn't appear to be a defining mutation for the parent clade. My haplotype is the parent clade of the one linked and my value is 146C.

First off, L3e2a1b(1/2/3) are as far as the haplogroups are defined officially. L3e2a1b1"a1b" is three branches down from that and is custom to yFull. So it's yFull's proprietary analysis that there was a back mutation downstream of you, and that it is relevant or branch-defining. The thing is that mito haplogroups and the science behind them are still in infancy (although the Million Mito Project will hopefully bring a lot of data to the table). FTDNA is based on the official standard, PhyloTree, which hasn't been touched in 6 years and is majorly out of date. As a result, it's the Wild West out there when it comes to project admins or yFull or whoever else suggesting new subclades.

According to this paper location 146 is unstable. It may actually be less reliable for defining haplogroups. (FTDNA currently only ignores 309 and 315, but there are many more locations that are potentially unreliable.)

You’re probably already aware, but FTDNA uses RSRS values over rCRS. FTDNA has their list of defining mutations here (click and scroll down), which should be virtually the same as on Phylotree: http://phylotree.org/tree/L3.htm Defining mutations for each branch build on each other, like with Y-DNA SNPs. L3e2a1b1 has its defining mutation, which is in addition to the one for L3e2a1b and the one for L3e2a1. Those are on top of the ones for L3e2a, L3e2, L3e, L3, and L1’2’3’4’5’6[‘7]. C146T is in the list for the very first L1–7 branch, so that mutation is baked into every haplogroup except for L0 descendants. If you have C at 146, that could imply that there was a back mutation in your lineage.

Your extra and missing mutations from your results should give you an idea of how much potential refinement there is in your own haplogroup assignment. There's a great walk-through for everything on FTDNA's results page.

[DJ Brooks]

Thank for your reply.