Skip to main content

Table 2 Empirical type I errors for n<100

From: MiRKAT-S: a community-level test of association between the microbiota and survival times

Number

Method

K w

K u

K 0.5

K BC

25

MiRKAT-S

0.054

0.055

0.055

0.056

 

Permutation

0.046

0.048

0.046

0.049

50

MiRKAT-S

0.045

0.058

0.051

0.051

 

Permutation

0.041

0.052

0.045

0.045

75

MiRKAT-S

0.054

0.058

0.051

0.053

 

Permutation

0.051

0.053

0.048

0.049

  1. Empirical type I errors for small sample sizes (n<100) with approximately 25% censoring. Results are based on 5000 simulated datasets, and permutation p values were obtained using 1000 permutations. K w, K u, K 0.5, and K BC represent results for the weighted UniFrac kernel, unweighted UniFrac kernel, Bray-Curtis kernel, and generalized UniFrac kernel with α=0.5, respectively