Microbiome Analysis: From Raw Reads to Publication

Conducted every alternate Sunday (10:00 AM – 5:00 PM IST · Live Online). Kindly register below and our team will get back to you with the upcoming cohort schedule and onboarding details.

Workshop Overview

This intensive, hands-on workshop walks you step by step through a complete microbiome data analysis workflow in R — starting from raw OTU/ASV abundance tables and taxonomy/metadata files all the way to statistically rigorous, journal-ready figures.

1. Introduction to R Packages Used in Microbiome Analysis

• Setting up reproducible RStudio projects and organizing ASV/OTU count matrices, taxonomy tables, phylogenetic trees, and sample metadata.

• Core ecosystem: phyloseq, microbiome, vegan, mia / TreeSummarizedExperiment, tidyverse, ComplexHeatmap, and ggplot2.

• Differential abundance frameworks: ANCOM-BC2, MaAsLin2, DESeq2, and ALDEx2.

2. Appropriate Data Filtering & Compositional Transformations

• Why microbiome count data are sparse, zero-inflated, and compositional — and how naive normalization leads to false positives.

• Data filtering best practices: sequencing depth cutoffs, prevalence thresholds (e.g., taxa present in ≥10% of samples), minimum relative abundance filtering, and contaminant removal.

• Choosing the right data transformation: Total Sum Scaling (TSS / relative abundance), Centered Log-Ratio (CLR), Robust CLR (rCLR), Hellinger transformation, log10(x + pseudocount), and when (or when not) to rarefy.

3. Alpha & Beta Diversity Matrices: Bray–Curtis vs. Euclidean Distance

• Alpha diversity indices: Observed richness, Chao1, Shannon entropy, Simpson index, and Pielou's evenness.

• Deep dive into Beta diversity distance matrices: Bray–Curtis dissimilarity (semi-metric, abundance-driven) vs. Euclidean distance & Aitchison distance (Euclidean distance on CLR-transformed compositional data), Jaccard presence/absence, and Weighted/Unweighted UniFrac.

• Ordination & hypothesis testing: PCoA, NMDS, PCA, PERMANOVA (vegan::adonis2), and dispersion testing (betadisper).

4. Selecting Appropriate Plotting Methods for Publication

• Heatmaps: Hierarchical clustering of top differentially abundant taxa and functional pathways with metadata annotation bars.

• Volcano Plots: Visualizing log2 fold-change against FDR-adjusted p-values to highlight statistically significant biomarker taxa.

• Box Plots & Violin Plots: Displaying group-wise alpha diversity and taxon abundance distributions with jittered individual points and statistical p-value brackets.

• Line Plots: Tracking longitudinal microbiome trajectories and treatment time-courses across cohorts.