Independent FF chromosome browsing

[Frederator]

FF's chromsome browser interface is not very useful. You have no reliable way to tell whether any 2 of your matches share a specific segment with each other.

Do something about this, please!!!

If nothing else you should be able to perform this type of independent browsing upon approving a relationship. It's an unreasonable burden to do it under the current set up, by tedious technical correspondence with people who don't understand the product as well as you do, and hence are likely to throw up their hands and drop the matter altogether.

[sjadelson]

Related, or perhaps a compromise if there are privacy issues involved here, might be a small matrix of relationships. Even if this is only 5x5 (like the Chrom Browser).

I'm a little fuzzy on the whole privacy line anyway. If I'm told that A is a match with a suggested level and I can see the individual chromosome match, and likewise for B, it's extremely handy to understand the relationship of A to B as well (which is Frederator's point, I think). Is the lack of such a feature one of privacy, technical ability, or just something feasible that hasn't made it to the top of the heap yet? I'm guessing either the first or last.

[Rebekah Canada]

Matches Matrix

Hello,

The team is looking at the Matches Matrix option with privacy concerns in mind. I will tell you more when they have something ready.

Related, or perhaps a compromise if there are privacy issues involved here, might be a small matrix of relationships. Even if this is only 5x5 (like the Chrom Browser).

[sjadelson]

Good to hear, thanks.

Side note: I appreciate the increased communication level in these forums, Rebekah.

[S9 H9]

Hello,

The team is looking at the Matches Matrix option with privacy concerns in mind. I will tell you more when they have something ready.

Under "Account Settings" a person already has the option to keep various information private. Could you just add another option, such as "allow your matches to do segment matching on your profile". Certainly you'd need to add a link explaining the whole idea for those people unaware of what it means, but it seems like something that could be opted in or out.

Being able to compare two of your DNA matches to each other is, I think, almost a prerequisite to doing the detailed kind of genealogical sleuthing many of us want to do on our ancestors.

[hansonrf]

Hello,

The team is looking at the Matches Matrix option with privacy concerns in mind. I will tell you more when they have something ready.

I might suggest they look at how 23andMe does it, once sharing established between users.

Bob

[KATM]

hansonrf, are you referring to Ancestry Tools/Family Inheritance: Advanced, or another function at 23andMe?

[hansonrf]

hansonrf, are you referring to Ancestry Tools/Family Inheritance: Advanced, or another function at 23andMe?

Yes, that one.

Once someone agrees to share, they essentially become part of your 'pool' of data sets. Your can then compare/contrast using the chromosome browser with anyone in your pool as the base person. Comparisons are limited to 3 at a time but can be saved to a spreadsheet or tables.

[Frederator]

I find the notion ridiculous

I find the notion that there is a serious privacy issue here utterly ridiculous.

It's understood that the segment matching algorithm is driven by a comparison of a mass of paternal AND maternal SNPs within statistical tolerances (i.e., mismatches, no-calls and breaks) that are arbitrarily selected. In fact, FTDNA's chromosome browser doesn't even identify those mismatches, no-call or break points, much less distinguish between the two contribution streams.

If we're really worried that someone is going to farm data on the vaguely defined start/end points of thousands of tiny segments and reconstruct a stranger's entire genome, we should stand back and ask ourselves if it wouldn't be easier to just pay someone to grab actual DNA from the door handle at the local mall.

[keithburgess]

I want to vote for this capability as well. If there is any privacy concerns, then you could just supply the amount of overlap that exists between the
2nd and 3rd person (in common pair) with my self and my match segments.
If answer is greater than 95%, then just report 95%. I don't care if you use
SNP counts or cM to determine % overlap.
All I care about is whether the match segment of the 2nd and 3rd person is
mostly overlapping my match segment and thus it is a match to that segment or not. As far as I am concerned, you could limit report to segments >=7.7 cM, You could even make this a selection parameter if you like.
There is no way to identify what the specific DNA sequence is of the third person match with an just an overlap value that can never be 100%