Skip to content

mitology

Dissect mitochondrial activity from transcriptome data — a Bioconductor package that scores RNA-seq samples against a curated reorganization of mitochondria-related Reactome and Gene Ontology pathways, plus pre-configured MitoCarta3.0 pathway sets, to quantify mitochondrial function from bulk or single-cell expression data.

  • Maintainer: Stefania Pirrotta (University of Padua) — stefania.pirrotta@unipd.it
  • Authors: Stefania Pirrotta (creator/author), Enrica Calura (author, funding); collaborator Massimo Bonora
  • Bioconductor: mitology v1.4.0
  • Source: git.bioconductor.org/packages/mitology
  • Vignettes: Introduction
  • License: AGPL-3
  • Status at GBCC2025: methods talk on the package's mitochondrial-activity scoring workflow

Talk at GBCC2025

  • Speaker: Stefania Pirrotta (University of Padua) — co-authors Massimo Bonora, Enrica Calura
  • Day / session: Day 4 (Thu Jun 26, 2025) — Oral Session 5
  • Talk title: "Dissect mitochondrial activity by transcriptome data with mitology R package"
  • Slides: to verify after publish
  • Video: GBCC2025 playlist
  • Abstract / schedule: GBCC2025 program

What it does

mitology takes an expression matrix (bulk or single-cell) and returns sample-level scores for mitochondria-centric pathways — OXPHOS, TCA, mitochondrial translation, fission/fusion, mitophagy, etc. — drawn from a curated reorganization of Reactome and GO terms plus the MitoCarta3.0 inventory. The package abstracts away the gene-set wrangling so an analyst can ask "which samples have low Complex I activity?" or "which clusters lean glycolytic vs. OXPHOS?" without hand-assembling pathway lists.

How it works

Core idea. mitology runs single-sample gene-set scoring (ssGSEA-like, per the vignette's "single-sample assessments" framing) against a curated mitochondrial pathway library, so each sample gets one score per mito gene set. The exact scoring algorithm is not specified in vignette beyond the single-sample framing.

Inputs / outputs. Input is a transcriptomic expression matrix (bulk or single-cell) with ENSEMBL IDs or gene symbols; SummarizedExperiment objects are supported. Output is a per-sample matrix of mitochondrial pathway activity scores, one row per gene set.

Key innovation. The gene-set library — mitology reorganizes Reactome and Gene Ontology pathways to focus on mitochondrial components and merges them with MitoCarta3.0, IMPI, MSeqDR, and GO mitochondrial inventories. This compensates for the way generic pathway databases tend to dilute mitochondrial signal inside broader cellular-signaling categories.

Parameters worth knowing. - Gene-set source — choice between MitoCarta3.0 pathways and the reorganized Reactome/GO sets. - Three-tier MitoCarta3.0 hierarchy — pathway organization level (broad function → process → specific complex). - Sample-level vs. group-level assessment mode. - Remaining tunables — not specified in vignette.

Canonical example. The vignette opens with the bundled gene-set database: 

library(mitology)
data(MitoGenesDB)
head(MitoGenesDB)   # ENSEMBL, SYMBOL, DB columns
Subsequent sections (truncated in the rendered vignette) chain this into a scoring call on an expression matrix to produce per-sample mitochondrial pathway scores.

Where it fits in the corpus

  • AACR 2026: axes = bioinfo-tools (mitochondrial metabolism is a recurring cancer-biology theme; mitology gives a transcriptome readout that complements metabolomics dossiers)
  • signifinder (entry) — same speaker (Pirrotta) and sister Bioc package; signifinder is the broader signature-scoring tool, mitology is the mitochondria-specialized variant of the same intellectual line
  • EuroBioC 2025: not presented
  • Nextflow Summit 2026: not mentioned

Notes

Pirrotta's GBCC slot pairs naturally with her signifinder work — both packages convert expression data to interpretable pathway/signature scores; mitology is the mito-focused descendant. The MitoCarta3.0 integration is the practical hook for cancer-metabolism users who already cite that resource.