A comparison of principal component methods between multiple phenotype regression and multiple SNP regression in genetic association studies (Q2179976)

This paper focuses on application and justification of the principal component analysis, which is the widely used method for dimension reduction in unsupervised multivariate analysis. The authors have demonstrated that the lower order principal components (with large eigenvalues) are generally preferred and lead to a higher power in setting for the single nucleotide polymorphism (SNP-set). Simulation studies as well as two real data analyses were employed to validate the findings.