Amplicon Week — Day 3: Visualization & Ecological Analysis Using QIIME2 + Microeco (R)

🌱 Amplicon Week — Day 3

Visualization & Ecological Analysis Using QIIME2 + Microeco (R)

After completing the QIIME2 core analysis (Day 2), today we move into the visualization and ecological interpretation phase using Microeco, one of the most flexible and modern R packages for microbial community data mining.

This workflow works for 16S, ITS, 18S, 12S, and COI — as long as you export your ASV table, taxonomy, tree, and metadata from QIIME2.

📦 What is Microeco?

Microeco is an R package designed for modular, flexible, comprehensive microbial ecological analysis using an R6 object-based framework.

⭐ Key features:

Creation of a unified microtable object

● Data cleaning & normalization

● Taxonomic visualization (barplots, heatmaps, composition)

● Venn diagrams

● Alpha & Beta diversity

● Differential abundance testing

● Co-occurrence network analysis

● Environmental data integration

● Supports multiple platforms: QIIME2, QIIME, HUMAnN, Kraken2, Phyloseq, MetaPhlAn, etc.

Microeco essentially brings the entire downstream microbial ecology workflow into one consistent R environment.

🧩 1. Creating a Microeco microtable object from QIIME2 outputs

Microeco expects the following input files:

After exporting in Day 2, you will have:

feature-table.tsv
taxonomy.tsv
tree.nwk
metadata.tsv

A. Option 1 — Using file2meco (recommended)

Microeco provides a companion package file2meco to directly read QIIME2 files into microtable.

install.packages("microeco")
devtools::install_github("ChiLiubio/file2meco")

library(file2meco)

mt <- file2meco::qiime2_to_meco(
  feature_table = "feature-table.tsv",
  taxonomy = "taxonomy.tsv",
  tree = "tree.nwk",
  metadata = "metadata.tsv"
)

This automatically:

● Cleans taxonomy

● Aligns sample names

● Creates the microtable object

B. Option 2 — Manual creation (phyloseq-style)

If you want full control:

library(microeco)
library(tidyverse)

otu <- read.table("feature-table.tsv", header = TRUE, row.names = 1, sep = "\t")
tax <- read.table("taxonomy.tsv", header = TRUE, row.names = 1, sep = "\t")
meta <- read.table("metadata.tsv", header = TRUE, row.names = 1, sep = "\t")
tree <- ape::read.tree("tree.nwk")

# Clean taxonomy (important!)
tax <- tidy_taxonomy(tax)

mt <- microtable$new(
  otu_table = otu,
  tax_table = tax,
  sample_table = meta,
  phylo_tree = tree
)

mt

🌈 2. Taxonomic Composition Analysis (Barplots)

Visualizing relative abundance across samples or groups. Create object:

t1 <- trans_abund$new(dataset = mt, taxrank = "Phylum", ntaxa = 20)

Plot:

t1$plot_bar(
  others_color = "grey70",
  facet = "Group",
  xtext_keep = FALSE,
  legend_text_italic = FALSE
)

✔ Shows top abundant phyla ✔ Automatically groups the rest as “Others” ✔ Facets by metadata (e.g., treatment, season, habitat) ✔ Can create same for top families, genus or species.

🟣 3. Venn Diagram (Shared ASVs/OTUs)

Useful for comparing core taxa across groups.

v1 <- trans_venn$new(dataset = mt, group = "Group")
v1$plot_venn()

📏 4 Diversity analysis

4.1 Microeco makes this very simple:

Microeco makes this very simple:

adiv <- trans_alpha$new(dataset = mt, group = "Group")
adiv$plot_alpha(boxplot = TRUE)

or individual metrics:

adiv$plot_metric("Shannon")
adiv$plot_metric("Chao1")

can be modified to include the ststistical analysis.

4.2 Beta Diversity (PCoA / NMDS)

Bray–Curtis PCoA:

bdiv <- trans_beta$new(dataset = mt, group = "Group", distmethod = "bray")
bdiv$plot_ordination(method = "PCoA")

NMDS example:

bdiv$plot_ordination(method = "NMDS")

Clustering patterns often reveal:

● treatment effects

● environmental gradients

● size fraction differences

● seasonal signatures

For incorporating ststistical analysis for diversity; please refer diversity_statistics.R in day3 of github project.

✅ WHAT THIS SCRIPT PRODUCES

Alpha diversity

● alpha_KruskalWallis_results.csv

● alpha_Wilcoxon_results.csv

● alpha_ANOVA_results.csv

Beta diversity

● beta_PERMANOVA_BrayCurtis.csv

● beta_Betadisper_BrayCurtis.csv

● Complete results object

● diversity_statistics.RData

📊 5. Differential Abundance

Microeco can run LEfSe-style analysis internally.

lf <- trans_diff$new(dataset = mt, group = "Group", method = "lefse")
lf$plot_diff_bar()

This identifies biomarkers at different taxonomic levels.

🔭 Microeco Can Do Even More

In this post, I focused on the core ecological analyses: composition, diversity, differential abundance, and basic networks. But Microeco can do much more, including:

● RDA / CCA and other ordination methods integrating environmental variables

● Correlation-based analyses (taxa–taxa, taxa–environment)

● WGCNA-style module detection and co-occurrence structure

● Functional prediction workflows and downstream visualization

For a deeper dive into these advanced features, I highly recommend the official tutorial: 👉 https://chiliubio.github.io/microeco_tutorial/

Keep in mind that many of these analyses are also possible directly in QIIME2. For more details on QIIME2-based visualizations and diversity analyses, see:

👉 qiime2_visualization.md in the Day 3 folder of the GitHub project https://github.com/jojyjohn28/AmpliconWeek_2025

If you are interested in trying Microeco, I’ve included all the required base files in my repository, along with a fully running R script. You can find: feature-table.tsv taxonomy.tsv tree.nwk metadata.txt and Dec10_microeco.R

These files allow you to reproduce the entire workflow exactly as shown.

Today’s image features visualizations I created myself using Microeco.

Thank you to the Campbell Lab’s graduate student seminar series — especially Dinu and Nichole — for the presentation on Microeco