region of where your genes came from because you'll be getting a random assortment of many, many genes from two conglomerated sources and the aggregate will be very similar. That's why we look so similar to siblings on a geographical description of gene sourcing.

My last comment was just responding to your example of any *particular* gene. If any one gene is not from your mom, then it is from your dad, and totally dependant on where the region was for that specific gene. And if your mom's geographical source for that particular gene is different from dad's (which is more likely than not), you will have a completely different region for this specific gene.

So in your example, if any particular gene is different from a sibling, it's most likely that it's NOT from the same region as hers for that particular gene. If your sister's gene is from Denmark, and she got it from your largely-Danish mother, and yours is not the same one as hers, yours is most likely not also from Denmark unless Dad happens to more than 50% Danish. It's going to be from wherever Dad got his instead.

When we look at the aggregate of all genes, the average is brought back because each variance comes from the same average geographical set. But for any individual gene it's either the same gene or it's not (~50% chance of either case) and if it's not, then it's less likely that mom and dad both got their version of that gene from the same place.

If you took my first comment to mean that the overall percentage of geographical source similarities would be very different, you're right to clarify. That wouldn't be the case. The genes would be ~50% different and if we looked at them specifically, they would show those differences. But if we are just mapping where they came from, the average would be very similar. Your specific example of "a gene" was the one that didn't make sense to me.