Skip to main content
Fig. 1 | Microbiome

Fig. 1

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

Fig. 1

Graphical illustration of the step-wise implementation of ZINQ. Step 1: Test of γ=0 by any valid test of logistic regression tells whether the variable of interest is associated with the presence-absence status of the taxon in samples. Step 2: Test of β(τk)=0 by the novel quantile rank-score test adjusting for zero inflation tells whether the variable of interest is associated with the difference at the τkth percentile of the taxon’s non-zero measurements. The testing is conducted marginally on K selected quantiles of the non-zero abundance, such as the default grid. Step 3: Combine the marginal p-values in Steps 1 and 2 considering the dependence structure of the tests. Only when the aggregate p-value is significant, we conclude that the taxon is differentially abundant

Back to article page