I notice that my two reports from competing DNA reports, Genographic 2.0 and 23andMe seemed to disagree with one another. Genographic 2.0 indicates 17% Southwest Asian, but 23andMe makes no mention whatsoever of any Asian. I assumed this confusion was due to a lack of understanding on my part, so I tweeted to the head of both projects to see if they could help clarify the results for me.
— Kevin Weatherly (@ikevinax) September 13, 2017
Luckily, Spencer Wells at (formerly at?) Genographic 2.0 replied almost immediately, referring me to a colleague of his, Razib Khan, who replied over three tweets.
using different methods and reference populations to make the judgment. way 2 think about is algorithm is asked question and returns result algorithms are being asked diff questions. basically depending on how you configure it northern europeans can have affinities with w asians or they do not. the underlying genetic patterns are the same, but the way u collapse variation for human presentation differs
It’s probably frustrating to understand a complex topic so thoroughly and not have a super simple way of conveying an idea to a layperson!