GraphExperiment: infrastructure for high-dimensional quantitative data and associated networks in R/Bioconductor

GraphExperiment extends SingleCellExperiment with an S4 class that stores feature-level and observation-level igraph networks in sync with assays and metadata, enabling coordinated graph–omics workflows.

• Why it matters: Bioconductor has interoperable data structures for quantitative omics, but lacks a standard container for networks alongside the data used to infer them.
• What they did: The package adds rowGraphs and colGraphs containers to an S4 class extending SingleCellExperiment, keeping igraph objects synchronized with assays and row/column metadata.
• The result: Across seven bulk and single-cell transcriptomics, proteomics, microbiome, and comparative genomics use cases, it supports manipulating diverse network types, from coexpression and co-abundance to cell communication and phylogenetic networks.

AI summary of the abstract · may contain errors · not medical advice · read the paper.