ZINQ_check: Sanity check before applying ZINQ

In wdl2459/ZINQ-v2: Zero-Inflated Quantile Approach for Microbiome Association Testing

Sanity check before applying ZINQ

Description

Sanity check before applying ZINQ

Usage

ZINQ_check(tax_tab, metadata, C)

Arguments

tax_tab	The taxa read count table (un-normalized), sample (row) by taxa (col).
metadata	The metadata, sample (row) by variable (col).
C	The name(s) of clinical variable(s) of interest, e.g., "Condition" or c("Condition", "Batch").

Details

• It is recommended to do the sanity check before applying ZINQ. If it is necessary, warnings will be printed to guide the analysis using ZINQ.

• If library size is a confounder of the variable(s) of interest, ZINQ might not control type I error.

• If there are few non-zero read counts, use ZINQ with caution.

• For a perfect separation, e.g., there are all zeroes in one group (case or control), which is an extreme case of imbalance, use ZINQ with caution.

• The sanity check is mainly about zero inflation. Most normalization methods will keep the original zeroes, thus investigating the un-normalized taxa read count table provides sufficient clues to use ZINQ. For normalization methods not retaining the zeroes. e.g., CLR, results of the sanity check is not informative, one can apply ZINQ directly.

Value

Print warnings if necessary:

• When library size is a confounder.

• For each taxon, (1) when all read counts are zero, (2) when there are limited non-zero read counts (<30 or <15), (3) when there is a perfect separation w.r.t. the variable(s) of interest.

wdl2459/ZINQ-v2 documentation built on March 25, 2024, 6:23 p.m.