Big Y - Results - Matching tab

[CRCampbell]

in the Big Y - results - matching tab... Is there any particular significance top the oreder the results are displayed when initially presented? Are they initially sorted Fewest Non-Matching Variants to Hightest number of Non-Matching Variants?

It would seem on first impression I would want to know which kits i share the higest number of matching variants with...and judge that against the non-matching variants?

"Number of matching, non-matching, and no call variants will total up to 923,438 for each pair of samples"...

Is there more significance to a lower number on Non-matching vs a higher number of matching? When I sorted high to low the Matching variants I could immediately see that my resreach apears to be on point. Thanks.

[CRCampbell]

Well, my current thoughts on this are... your matches are sorted first by lowest number of non-matching variants to show which kits are closest to yours using known SNPs and private variants. So, it follows, the more non-matching variants plus private variants, the further away from your own Terminal SNP that kit will be. Knowing that number would also let you play around with SNP based dating estimates... Last I heard two years ago or so allegedly FTDNA was giving a range of 82 to 98 years per SNP. That seems over-simplified ...I would think you'd want to count down from the common SNP? But whatever...

The problem with sorting by Shared variants is the number of no-calls per kit...your own and the one being compared.... The number will be different for each kit and it would be difficult to tell where the overlap of those no-calls is at for each and every kit.... Would the result be because you had few no-calls and they had alot? Vice versa? You both had alot at the same positions? You both had few no-calls but non-of them were at the same positions so it looks like alot, but if we were able to get reads on all those it may actually be very few...too many possiblities.

So, to answer myself...it would appear that its sorted by fewest Non-matching variants because its based on actual reads that could be compared.

What I'll end up doing is looking at the Most shared, say the top ten, and compare that to where they are on the Non-matching list....the idea being that the higher the Shared and the lower the Non-matching the better the target.... Caveat..I'm referring to recent BigY700 results...with mych more recnt SNPs being discovered and how that might be used in the coming years...

[randalgibbs]

I'd ignore the "shared variants" column. The "shared variants" of newer matches will be higher as FTDNA is continually finding new variants to compare. Currently the number of variants considered for matching = 929,977. That number will increase over time, so future matches to your kit will have more "shared variants" than your current matches do. (They don't re-calculate the number of shared variants to your existing matches as new variants are discovered).

[CRCampbell]

I'd ignore the "shared variants" column. The "shared variants" of newer matches will be higher as FTDNA is continually finding new variants to compare. Currently the number of variants considered for matching = 929,977. That number will increase over time, so future matches to your kit will have more "shared variants" than your current matches do. (They don't re-calculate the number of shared variants to your existing matches as new variants are discovered).

Ah, thanks! I keep forgetting there are older kits mixed in...hmmm.