This repo provides the code and example data to run ZINQ-L.
We propose a zero-inflated two-part quantile regression approach for differential abundance (DA) analysis of longitudinal microbiome data (ZINQ-L)
To run the code, you need to clone the repo.
git clone https://github.com/stephlee3/ZINQ_L.git
cd ZINQ_L
Install and load all the necessary packages.
library(tidyverse)
library(phyloseq)
library(GUniFrac)
library(quantreg)
library(parallel)
library(lmerTest)
library(lme4)
library(CompQuadForm)
library(expm)
library(FactoMineR)
library(Matrix)
library(MASS)
library(LDM)
library(MicrobiomeStat)
library(Maaslin2)
library(NBZIMM)
Now you can run the example code in example.R
. A matrix of P-values (each row: a replicate of representative taxa, each column: a statistical test methodology) will be generated on the simulated data.
This repo is organized as follows
-
Code/
util.R
: contains all the statistical tests.README.md
: description of tests.
-
example.R
: the example code to run the tests on simulated data. To run the sim, please contact the author to access the required data.
For any additional questions, please contact Runzhe Li([email protected]).