Skip to main content
. 2016 Feb 10;24(9):1344–1351. doi: 10.1038/ejhg.2016.8

Table 2. Type I error comparison between MURAT and SKAT at stringent significance levels, α=1 × 10−4, 1 × 10−5, and 2.5 × 10−6.

  ρe=0 ρe=0.4
α MURAT SKAT MURAT SKAT
K=2
 1 × 10−4 1.68 × 10−4 1.27 × 10−4 1.48 × 10−4 1.24 × 10−4
 1 × 10−5 2.78 × 10−5 1.16 × 10−5 1.29 × 10−5 1.54 × 10−5
 2.5 × 10−6 7.60 × 10−6 3.30 × 10−6 2.80 × 10−6 3.30 × 10−6
         
K=3
 1 × 10−4 1.64 × 10−4 1.29 × 10−4 1.35 × 10−4 1.23 × 10−4
 1 × 10−5 2.15 × 10−5 1.25 × 10−5 1.82 × 10−5 1.35 × 10−5
 2.5 × 10−6 9.80 × 10−6 4.10 × 10−6 10.20 × 10−6 3.10 × 10−6

Abbreviations: MURAT, Multivariate Rare-Variant Association Test; SKAT, sequence kernel association test.

The simulations are based on 107 simulated data sets. The results for SKAT are based on adjusted P-values, which are defined as K times the minimum of univariate-based P-values obtained via SKAT.