Louis Kessler’s Behold Blog The Behold User Forum
Blog Comments
72.
The GEDCOM 5.5.5 Initiative and Making It Work - Blog comment by metti - 27 Oct 2019
I wish there were a GEDCOM version, witch clear some issues and spend more possibilities.
The specification 5.5.5 from Tamura Jones looks like a good step in the right way. For me there are too many restrictions in it. I can't export all of my user-data. So the user loose informations if my program will use ...
73.
23andMe’s Family Tree Beta - Blog comment by lkessler - 25 Sep 2019
Keith: No, it's not under "DNA Relatives". Click on the "Family & Friends" menu item and you should see it.
74.
23andMe’s Family Tree Beta - Blog comment by geneatech - 25 Sep 2019
Louis, I read on Reddit that the tree is shown only for people you have connections to—is that your experience? I'm in the beta program but don't see the family tree anywhere (I believe it's supposed to be under DNA Relatives, right?).
75.
The Life and Death of a DNA Segment - Blog comment by lkessler - 22 Aug 2019
Ahh. Now I see what you're trying to do. You're taking the two pieces of the 15 cM segment, and deciding that if one was only 5 cM, then it wouldn't match anyone because the smaller side wouldn't reach the threshold that the company calls their minimum match.Yes. That's entirely possible and may be the reason ...
76.
The Life and Death of a DNA Segment - Blog comment by jonathanb - 21 Aug 2019
I guess I'm hung up on "the 15 cM segment". From what I've seen, it's rare that three people share the EXACT same segment, with exact same start and end points. So when I see "the 15 cM segment", I read "a 15 cM region of DNA, where various matches might share various overlapping bits within that ...
77.
The Life and Death of a DNA Segment - Blog comment by lkessler - 21 Aug 2019
Jonathan: I don't understand your logic why a 6 cM limit means that the 15 cM wasn't born in some ancestor, but got longer. It's always a recombination that creates a new segment. Whether it or parts of it are filtered out by the company doesn't seem relevant to me because the filtering is only done in the DNA ...
78.
The Life and Death of a DNA Segment - Blog comment by jonathanb - 20 Aug 2019
Nice writeup, I've been waiting for someone to talk about this.
There's a corollary that you didn't talk about. Ancestry has a 6 cM minimum size for matching segments. Any segment under 6 cM might as well not exist, as far as Ancestry's matching algorithms are concerned. Other sites use slightly different ...
79.
Comparing Raw Data from 5 DNA Testing Companies - Blog comment by lkessler - 18 Aug 2019
Yinwang: The general method of imputation is to use samples from a population that best fit known values to fill in missing values. Imputed values will thus not catch many variants, and may result in non-matching segments with close relatives who should match and matching segments with people who shouldn’t ...
80.
Comparing Raw Data from 5 DNA Testing Companies - Blog comment by yinwang888 - 17 Aug 2019
This is super interesting and well done. I think ultimately the double position and double rs-ids problem are to blame on dbSnp the underlying public database. I have heard many professional geneticists say that they now only use chr:pos:A1_A2:built as ID because of this frustration (that’s “chromosome” , ...
Prev 61-70
The GEDCOM 5.5.5 Initiative and Making It Work - Blog comment by lkessler - 27 Oct 2019