Skip to main content

Table 2 Type I error control and power on simulated data based on Anaerovorax’s normalized abundance

From: Powerful and robust non-parametric association testing for microbiome data via a zero-inflated quantile approach (ZINQ)

Sample size = 600      
  Type I error Power
  Null Setting 1 Setting 2 Setting 3
α-level 0.05 0.01 0.05 0.05 0.05
Rarefaction      
Linear regression 0.0547 0.0084 0.9949 0.6928 0.1247
ZIP 0.7387 0.6622 1.0000 0.9742 0.7720
ZINB 0.2019 0.0771 0.9812 0.7321 0.2398
ZINQ-MinP 0.0526 0.0106 0.9994 0.8557 0.1508
ZINQ-Cauchy 0.0580 0.0110 0.9991 0.8346 0.1493
TSS      
Linear regression 0.0536 0.0088 0.9970 0.7425 0.1320
ZIB 0.0110 0.0017 0.9964 + 0.6255 + 0.0305 +
Tobit 0.0543 0.0099 0.9989 0.8041 0.1467
ZIlogN 0.6992 0.6872 1.0000 1.0000 0.9999
ZIG 0.0548 0.0102 0.9961 0.7264 0.1196
ZINQ-MinP 0.0501 0.0101 0.9995 0.9096 0.1669
ZINQ-Cauchy 0.0503 0.0103 0.9994 0.8981 0.1555
CSS      
Linear regression 0.0527 0.0113 0.9995 0.8934 0.1733
Tobit 0.0526 0.0110 0.9985 0.8597 0.1628
ZIlogN 0.0475 0.0095 0.9996 0.8794 0.1464
ZIG 0.0494 0.0096 0.9998 0.8850 0.1474
ZINQ-MinP 0.0501 0.0103 0.9993 0.8852 0.1520
ZINQ-Cauchy 0.0505 0.0095 0.9991 0.8735 0.1524
  1. Setting 1: 100% from HBP edf for HBP samples;
  2. Setting 2: 80% from HBP edf and 20% from non-HBP edf for HBP samples;
  3. Setting 3: 60% from HBP edf and 40% from non-HBP edf for HBP samples.
  4. : power of a method that inflates type I error
  5. +: power of a method that deflates type I error
  6. Results by the various methods on 10000 simulated datasets by generating samples from the edf of Anaerovorax’s normalized abundance, including type I error control and power under different settings with significance cutoffs 0.05 and 0.01