Tree.Ratio
Differential Abundance Analysis for Microbiome data Incorporating Phylogeny
Installation
# Install the released version from CRAN
install.packages("LTRR")
# Install the development version from GitHub
devtools::install_github("ZRChao/LTRR")Contents
Simulation
- BIT.Sim.R for Multinomial(Binomial) Tree distribution
- DTM.Sim.R for Dirichlet Multinomial Tree distribution
- LNM.Sim.R for Logistical Normal Multinomial distribution
- ANCOM.Sim.R for Poission distribution (parameters set follow as ANCOM paper)
Tree relate function
- Tree.Sim.R : Tree simulation
- Taxa.index.R : relationship between leafs and internal nodes
- Prob.mult.R :calculate the probability of each leafs by multiple each probability along the branch
- Diff.otu.R : judge the probability of each node in two group equally or not
- Tree.ratio.R : Tree ratio test based on the tree structure
Other function
- pow.fdr.R : calculate power and fdr with p.value
- Zig.pv.adj.R : metagenomeSeq::fitFeature
- ancom.py : which is faster in python
Require
- ancom.R::ANCOM, metagenomeSeq::fitFeature, mvtnorm, gtools, rdirimult
Example
#You can see this in real data application of throat.R
library(MiSPU)
load(throat.otu.tab)
load(throat.tree)
load(throat.meta)
p = ncol(throat.otu.tab)
throat.taxa.index <- Taxa.index(p, throat.tree)
throat.alltab <- cbind(throat.taxa.index*throat.otu.tab, throat.otu.tab)
group <- throat.meta$SmokingState)
result <- Tree.Ratio(p, throat.tree, throat.taxa.index, throat.alltab, group)
throat.detected <- Tree.Ratio.back(p, throat.tree, throat.taxa.index, results, group)