Matches and Chromosome Browser

[jojay]

I’d be grateful if someone would explain to me the point of the following scenario: from my Autosomal Match list I check the box beside A Smith then click for it to show me who I have in common with him. This brings up a list of matches. I select B Brown so that they can be compared in the Chromosome Browser. After clicking on Compare Relationship The list of chromosomes with their coloured segments appears. None are shared. I know that I can only view segments of 6cM or more but if they don’t meet the criteria, why tell me? I’ve tried this on both laptop and iPad and it’s the same answer. It’s not like this for everyone I try to compare, just some. Please someone, shed some light. Thanks.

[georgian1950]

A common match may match you via one ancestor, but match the person selected for the common matches via another ancestor. Or it could be that all have the same common ancestor, but the matches are on a different part of the genome. One just can't say for sure without a triangulated match. As far as I am concerned, FTDNA's new 6 cM minimum matching segment size has made their Family Finder product useless for me as it throws out small segments that might prove a common ancestor.

Here are some thought of mine about short segment matching:

Genetic Genealogy Out of the Mainstream – questioning the assumptions, examining the credibility

[Gwailo]

I agree georgian1950 that it was a backward move removing those matches with small segments for those of us wanting to reach back beyond the obvious connectivity of the past 150 years or so. I lost matches some of who were 50+ cm total with longest block of 9cm which were matching some others who I still match now in the latest version. This is especially important where you have parents from quite different parts of Europe or some other place because those small segments still highlight easily what regional association those matches have via the in common function.
I hope they bring them back soon because now when someone matches me at 12cm total shared it actually could be closer to 60cm which is a not insignificant number for most of us looking further back than 3 or 4 gens ( which I would have thought is the whole point of genetic ancestry testing - to look back further than the people you already know where they're from)

[georgian1950]

Gwailo, I appreciate you taking a look and posting your comment. I have been a bit half-hearted about the blog as you can tell by the length between posts. I am frustrated enough with what has happened with autosomal matching that I intend to pick up the pace with the blog, I hope to have another post out soon.

I anyone likes what I wrote, please share with anyone else who may be interested

[Gwailo]

Good on you. If you are on twitter it's probably not a bad idea to share to your blog from there also. You should also think of contacting Göran Runfeldt, Genetic Genealogist from Sweden, Head of R&D at ftdna to further convince him See him in this video talking with Sturla talking about ancient origins development tools they are working on > https://www.youtube.com/watch?v=krmKE1Jd49Y

[georgian1950]

Appreciate the suggestions. I am not big on social media so I haven't tried Twitter yet, but probably should. I am on Facebook but don't do a lot there. I do show my blog on my profile there. I've been active on a surname group, so I think that I am going to push the link out to them in the next few days and see if I get any interest.

Unfortunately, the DNA companies and services like GedMatch will not restore small segments unless users demand it. Right now some big names in genetic genealogy are holding too much sway. They have tried to make logical arguments against short segments, but use bad logic as the counter-examples in my last blog entry show. Why not just simple confidence levels based on probability? The 100 SNP match meets that standard with flying colors. One thing people don't see is that frequently a match is built up by having more than one line to the common ancestor. However that is a problem with the distant big segment matches, too. It is likely more of a problem with these so called IBD segment matches, than it is with shorter segments.

[Gwailo]

You're welcome & I agree with you about some of the larger IBD segments. I have one that appears to be inherited across Scandinavia & Scotland at almost the exact same start & end positions. Its unbelievable to see almost 100 people inheriting a block that is almost identical to the exact positions and amount of SNPs. But with these as with other segments , your so called "REAL" relatives will match you on these closer in time as well as those who are connected further back. I think these new policies are to do with limiting their database size / server utilisation from a cost IT infrastructure POV more than to help us solve our family mysteries. I will keep pushing for them to widen the net too. Thanks & keep up the good work