Pile-up Regions is a name for some regions of DNA on some Chromosomes where there is not much variation. This leads to false segment matching with those you likely do not have a genetic nor genealogical relationship with. The lack of variation can be due to a number of issues. Fewer SNPs tested in the range. Less recombination happening in that area (leading to ancient segments of DNA persisting among members of the population). Or others. Key is to understand that when looking at small, single segment matches (less than say 30cM in size), the likelihood of a false match is higher. Focus on your larger matching segments and total matching segment connections first.
Note: some attribute false matches to phased versus un-phased information. Phasing simply tries to recreate the contribution of an allele (SNP value) from one parent versus the other. That is, in the un-ordered pair of SNP values returned on each chromosome, phasing tries to bring order or identify the source of that SNP. The thought is that the half-identical match tools are creating many extra false matches by aggressively creating the longest matching segments possible using either value of the pair. The tools cannot do anything different. This is an artifact of the testing process and trying to make useful tools with what data is available. They further go on to show how phased results drop many of the smaller-length matching segments. But there are a number of issues going on here. Both with the phasing tool / process and the underlying data. If you have both parents tested, go ahead and phase your results. Then do a detailed comparison of your phased result from each side with each parent. You may be surprised.
Note: some attribute false matches to phased versus un-phased information. Phasing simply tries to recreate the contribution of an allele (SNP value) from one parent versus the other. That is, in the un-ordered pair of SNP values returned on each chromosome, phasing tries to bring order or identify the source of that SNP. The thought is that the half-identical match tools are creating many extra false matches by aggressively creating the longest matching segments possible using either value of the pair. The tools cannot do anything different. This is an artifact of the testing process and trying to make useful tools with what data is available. They further go on to show how phased results drop many of the smaller-length matching segments. But there are a number of issues going on here. Both with the phasing tool / process and the underlying data. If you have both parents tested, go ahead and phase your results. Then do a detailed comparison of your phased result from each side with each parent. You may be surprised.
External References
- Khazaria site analysis
- Jim Bartlett's Blog Post on pile-ups
- Debbie Kennett's pile-up visualization blog post using an experimental tool by Dan Edwards
- Sue's blog post of her map of problematic regions
- Identical by Descent on ISOGG that covers some analysis of false matching (see section on Excess IBD)
- AlternatingDNA chart (visual) to understand how half-identical match technology on standard autosomal testing generates false or extended segments