Power of ADMXITURE mapping

Question

[GabrielMontenegro]

I am conducting an admixture mapping association. Before that I also conducted a standard genotype-association. I want to compare my results between the two tests, so I was trying to find a paper that would tell me how to estimate the power of an ADMIXTURE mapping analysis.

For instance, I know that this will depend on the allele frequency differences between the populations and the difference in means or incidence of the trait in each ancestral population, but I was looking for a formula.

I am looking for something like this: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1181990/pdf/AJHGv74p979.pdf

Any other similar papers?

Answer 1

[Petr Ponomarenko]

I am not sure what do you mean with ADMIXTRURE mapping, but whenever we use ADMIXTURE and admixture vectors for analysis we create our own statistic test that is based on these vectors as inputs. We have a set of true and mockup populations from around the world with known and modeled characteristics. We use that set of populations within the same ADMIXTURE run and we find the power of our statistic on that data. Sometimes part of this data is used for training if our statistic test that based on admixture vectors has a parameter that needs to be optimized for a specie, super population, geographic region, etc.

[GabrielMontenegro]

The link I provided with my question gives an example of what I mean by admixture mapping. This is the first result I get when googling ADMIXTURE MAPPING: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3556814/