tidyMicro: a pipeline for microbiome data analysis and visualization using the tidyverse in R

Area of Science:

• Microbiology
• Bioinformatics
• Computational Biology

Background:

• Microbial communities interact with environments, driving innovation across fields.
• High-dimensional sequence-based microbiome data present challenges in merging and analysis.
• Existing R packages have limitations in data structure familiarity and workflow flexibility.

Purpose of the Study:

• To develop a comprehensive microbiome analysis pipeline in R.
• To provide tools for managing high-dimensional microbiome data and metadata.
• To offer enhanced analyses beyond community-level and exploratory visualizations.

Main Methods:

• Developed the open-source R package "tidyMicro".
• Integrated standard microbiome analysis tools (e.g., count table management, diversity inference).
• Incorporated advanced regression modeling (negative binomial, beta binomial, rank-based testing) and novel visualizations.

Main Results:

• tidyMicro offers essential tools comparable to existing popular packages.
• The package includes novel visualizations like Rocky Mountain plots and longitudinal ordination plots.
• Maintains familiar R data structures for improved user control and workflow flexibility.

Conclusions:

• tidyMicro serves as a reliable alternative for microbiome analysis in R.
• It extends standard analyses with novel approaches to enhance interpretability.
• The package is reproducible, applicable to external datasets, and encourages open-source collaboration.