Skip to content

mia

GBCC2025 workshop / oral pair on the Bioconductor microbiome ecosystem. See EuroBioC 2025 entry for the same package's parallel coverage at the European event.

Microbiome Analysis with Bioconductor — a TreeSummarizedExperiment-based framework for microbiome data, providing diversity, ordination, differential abundance, and community-ecology analyses on a single canonical S4 container. The substrate for the OMA (Orchestrating Microbiome Analysis) book and the mia* family of helper packages.

  • Maintainer: Tuomas Borman (University of Turku) — tuomas.vesa.borman@utu.fi
  • Authors include: Felix G.M. Ernst, Tuomas Borman, Leo Lahti (PI), and broader microbiome community
  • Bioconductor: mia v1.20.0
  • Source: github.com/microbiome/mia
  • Companion book: Orchestrating Microbiome Analysis (OMA)
  • License: Artistic-2.0
  • Status at GBCC2025: mature package; workshop + Oral Session 5 talk

Workshop + talk at GBCC2025

  • Workshop presenter: Tuomas Borman (University of Turku)
  • Workshop session: Day 3 — afternoon workshop slot (2-3:30 PM)
  • Workshop title: "Orchestrating Microbiome Analysis with Bioconductor"
  • Oral speaker: Leo Lahti (University of Turku)
  • Oral session: Day 4 — Oral Session 5
  • Format: hands-on workshop drawing on the OMA book; companion oral talk on ecosystem direction
  • Slides / materials: to verify after publish — typically a per-event GitHub repo under github.com/microbiome
  • Video: GBCC2025 playlist
  • Abstract / schedule: GBCC2025 program

What it does

mia gives microbiome data a TreeSummarizedExperiment home — abundances, taxonomy, phylogenetic tree, and per-sample metadata in one S4 object — and supplies the standard analysis vocabulary (alpha/beta diversity, ordination, agglomeration to taxonomic ranks, differential abundance via several backends, transformation utilities). The mia* family extends this with time-series (miaTime), visualization (miaViz), simulation (miaSim), and Shiny GUIs (iSEEtree). The OMA book is the curated walkthrough.

How it works

Core idea. mia treats the microbiome as a single TreeSummarizedExperiment object that keeps abundance assays, taxonomy rowData (Kingdom → Species), phylogeny in rowTree, and sample metadata in colData synchronized — so subsetting and analysis propagate through every layer at once.

Inputs / outputs. Inputs are imported from BIOM files, QIIME2 output, DADA2 objects, or phyloseq objects via the mia importers. Outputs include alpha- and beta-diversity values (Shannon, Simpson, JSD, UniFrac, …), agglomerated assays at a chosen taxonomic rank, transformed assays (log, rarefaction), differential-abundance results, and tidy data frames for downstream plotting.

Key innovation. A single S4 container that unifies microbiome operations — abundance, taxonomy, and phylogeny stay aligned under the standard SummarizedExperiment idioms — which makes mia the substrate for the OMA book and the broader mia* family (miaTime, miaViz, miaSim, iSEEtree).

Parameters worth knowing. - agglomerateByRank() — collapse features to a taxonomic rank (e.g. rank = "Family"). - addAlpha() — compute and attach alpha-diversity indices (index = "shannon", etc.) to colData. - transformAssay() — apply a transformation (log, relative abundance, CLR, …) and store it as a new assay. - rarefyAssay() — equalize sequencing depth across samples.

Canonical example. The vignette loads the bundled GlobalPatterns dataset and agglomerates 19,216 OTUs down to 341 families, then attaches Shannon diversity in one step: 

data(GlobalPatterns, package = "mia")
tse <- agglomerateByRank(GlobalPatterns, rank = "Family")
tse <- addAlpha(tse, index = "shannon")

Where it fits in the corpus

  • AACR 2026: axes = bioinfo-tools (microbiome–cancer interfaces appear in immune-microenvironment and cancer-prevention dossiers)
  • miaTime (entry) — same UTU group; longitudinal extension built on mia's containers
  • Bioc-to-Galaxy (entry) — Borman is core to both the microbiome ecosystem AND the Galaxy-bridge CoFest; mia is a flagship candidate for auto-integration
  • EuroBioC 2025 mia entry (entry) — the European twin
  • Nextflow Summit 2026: not mentioned

Notes

The GBCC angle on mia emphasizes Galaxy / cross-platform integration — the joint Galaxy-Bioconductor framing of the event makes "how do non-R users get at this" a first-class question. The EuroBioC version was Bioc-internal (S4 design, OMA book curation, mia* family extensions); GBCC is the cross-ecosystem export. Both tracks share the same package and team — the framing is what differs.