Statistical methods for microbiome data analysis
Pratheepa Jeganathan,Susan Holmes McMaster University
Abstract
This workshop will enable users to follow the workflow for reproducible research on statistical methods in molecular microbial data analysis published in \href{https://arxiv.org/abs/2103.04198}{Jeganathan and Holmes (2021)}. We will demonstrate a few statistical tools that can overcome strain switching in molecular microbial data. An important goal is often to find taxonomic differences across environments or groups, strain switching can be an impediment, and we introduce differential topic analysis that facilitates inferences on latent microbial communities.
In this workshop, the modules are the goodness of fit tests, power analysis, topic analysis, and differential topic analyses. We will show how to use topic models to provide useful aggregates for differential abundance analysis based on topics rather than individual strains using an R package diffTop available on Github. The materials for the workshop are available at \href{https://pratheepaj.github.io/diffTop/}{https://pratheepaj.github.io/diffTop/}.
Keywords: Microbial ecology,Bayesian data analysis,hierarchical mixture models,latent Dirichlet allocation,differential abundance analysis,sequencing data