Understanding the population structure correction regression

The Tien Mai,Pierre Alquier,Tien Mai

DOI: https://doi.org/10.48550/arXiv.2108.05655

2021-08-12

Applications

Abstract:Although genome-wide association studies (GWAS) on complex traits have achieved great successes, the current leading GWAS approaches simply perform to test each genotype-phenotype association separately for each genetic variant. Curiously, the statistical properties for using these approaches is not known when a joint model for the whole genetic variants is considered. Here we advance in GWAS in understanding the statistical properties of the "population structure correction" (PSC) approach, a standard univariate approach in GWAS. We further propose and analyse a correction to the PSC approach, termed as "corrected population correction" (CPC). Together with the theoretical results, numerical simulations show that CPC is always comparable or better than PSC, with a dramatic improvement in some special cases.