PDA

View Full Version : Gedmatch phased kit vs unphased kit and probability of IBD?



therrien.joel
10-21-2017, 06:16 PM
I've had the opportunity to check my Gedmatch kit against someone who has their phased. The majority of matching segments remain mostly unchanged between the two. Now, what I want to know is if anyone has worked out how the probability of a match being IBD vs match length in cM (or SNPs) changes when using a phased kit. I'm thinking something similar to what I believe I saw on the ISOGG website.

Donwulff
10-21-2017, 08:26 PM
I've done some experiments with HRC 1.1 imputation and phasing; other people may be phased against both parents which would be the only way to get proper phasing. And of course then there's DNA.Land phased kits. So the answer will depend somewhat on which phasing is in question. I don't think the probabilities people have been calculating take phasing into account at all though; beyond certain cM match length it starts to become a non-issue, and the matches below that segment are likely background sharing in population anyway. It's an interesting question to explore, though.

therrien.joel
10-22-2017, 02:09 AM
Yes, in this case I was referring to parental phasing.