Briefly, phyloseq takes in data from data processing programs like QIIME, mothur, and Pyrotagger. While QIIME2 offers richness estimates and other exploratory data analysis (ex: alpha and beta diversity metrices) we believe that phyloseq in combination with ggplot2 offers greater flexibility for generating customizable data visualizations. Once exploratory data analysis in phyloseq is complete, we use the LDM package to perform statistical analyses. LDM takes in a table of operational taxonomic units (OTUs) or amplicon sequence variants (ASVs) along with a table of data about the samples (i.e. covariates) and uses a linear decomposition model to associate experimental conditions and covariates of interest with microbial abundance. Data about the samples typically includes sample names, some experimental condition of interest, and other variables as collected by the experimentors. LDM can accomodate both continuous and categorical data.

Briefly, phyloseq takes in data from data processing programs like QIIME, mothur, and Pyrotagger. There are many other ways to import data into phyloseq. We typically import the .qza files produced by QIIME2 but in this walkthrough we will be using a built-in dataset that you can use anytime so you can follow along with this walkthrough if desired. For more examples to import data using other programs, see the phyloseq vignette here https://bioconductor.org/packages/release/bioc/vignettes/phyloseq/inst/doc/phyloseq-basics.html While QIIME2 offers richness estimates and other exploratory data analysis (ex: alpha and beta diversity metrices) we believe that phyloseq in combination with ggplot2 offers greater flexibility for generating customizable data visualizations. Once exploratory data analysis in phyloseq is complete, we use the LDM package to perform statistical analyses. LDM takes in a table of operational taxonomic units (OTUs) or amplicon sequence variants (ASVs) along with a table of data about the samples (i.e. covariates) and uses a linear decomposition model to associate experimental conditions and covariates of interest with microbial abundance. Data about the samples typically includes sample names, some experimental condition of interest, and other variables as collected by the experimentors. LDM can accomodate both continuous and categorical data.