Visualization and statistical comparisons of microbial communities using R packages on Phylochip data

Area of Science:

• Microbiology
• Bioinformatics
• Statistical Analysis

Background:

• Irritable Bowel Syndrome (IBS) is a complex gastrointestinal disorder.
• Gut microbiome alterations are implicated in IBS pathogenesis.
• High-throughput sequencing technologies like the LNBL Phylochip enable comprehensive microbiome analysis.

Purpose of the Study:

• To present a robust statistical and computational methodology for analyzing species abundance data from the LNBL Phylochip.
• To identify microbial species associated with Irritable Bowel Syndrome (IBS) in a rat model.
• To adapt and apply existing microarray analysis tools for microbiome data.

Main Methods:

• Utilized Family Wise Error rate control and step-down tests for multiple testing correction.
• Employed hypergeometric tests for identifying significant phyla and families.
• Applied normalization, multivariate projections, and batch effect detection techniques.
• Integrated phylogenetic covariation, tree equalization, and robustification methods.

Main Results:

• Successfully applied established statistical tools for analyzing complex microbiome datasets.
• Demonstrated a framework for identifying key microbial taxa associated with IBS in rats.
• Provided practical examples of data processing and analysis steps.

Conclusions:

• The presented methodology offers a powerful approach for microbiome data analysis in IBS research.
• This framework facilitates the discovery of microbial biomarkers for IBS.
• The study highlights the utility of adapting existing bioinformatics tools for novel biological questions.