Abstract

Testing gene-based associations is the fundamental approach to identify genetic associations in sequencing studies. It is also commonly used in other genetic association studies as an effective yet biologically meaningful way to enhance the statistical power. The best-known approaches include Burden and Sequence Kernel Association Tests (SKAT). The gene-traits associations are often complex due to population heterogeneity, gene-environmental interactions, and various other reasons. The mean-based tests, including Burden and SKAT, may miss or underestimate some high-order associations that could be scientifically interesting.

In this paper, we propose a new family of gene-level association tests, which integrate quantile rank score processes while combining multiple weighting schemes to accommodate complex associations. The resulting test statistics enjoy multiple advantages. They are as efficient as the mean-based SKAT and Burden test when the associations are homogeneous across quantile levels and have improved efficiency for complex and heterogeneous associations. The test statistics are distribution-free, and could hence accommodate a wide range of distributions. They are also computationally feasible. We established the asymptotic properties of the proposed tests under the null and alternative hypothesis and conducted large scale simulation studies to investigate its finite sample performance. We applied the proposed tests to Metabochip data to identify genetic associations with lipid traits and compared the results with those of the Burden and SKAT tests

For more information about Tianying visit:  
https://tianyingw.github.io/