CorNetto

CorNetto is a Bioconductor-style R package for knowledge-guided multi-omic correlation network analysis. It is designed for normalized transcriptomic, proteomic, and metabolomic abundance data stored in a MultiAssayExperiment.

The first package version focuses on:

• group-specific correlation networks
• pre-analysis summaries of sample counts, missingness, and feature counts
• two-group differential correlation testing, over all pairs in an assay or over candidate edges taken from a prior-knowledge network
• rewiring scores for perturbed nodes
• permutation-based assessment of rewiring scores
• pathway-focused subnetworks
• graph construction and Cytoscape-ready export

Workflow

Green boxes are functions, blue boxes are the objects they return, and orange boxes are inputs and the MultiAssayExperiment that carries results between steps.

Installation

CorNetto is under review for Bioconductor. Once accepted, install it with:

if (!require("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

BiocManager::install("CorNetto")

Until then, the development version can be installed from GitHub:

BiocManager::install("bradleyalexward/CorNetto")

Getting started

library(CorNetto)

analysisData <- exampleAnalysisData()
knowledgeNetwork <- exampleKnowledgeNetwork()

# With assayName omitted, every assay is tested densely within-omic.
# Protein-protein and transcript-transcript pairs are included, for example,
# but protein-transcript pairs are not generated.
differentialResults <- testDifferentialCorrelation(
    analysisData,
    groupColumn = "clinicalGroup",
    groupLevels = c("Recovered", "PASC"),
    storeResult = FALSE
)

differentialNetwork <- createDifferentialCorrelationNetwork(differentialResults)
rewiringScores <- calculateRewiringScores(differentialNetwork)

Long permutation runs can use a BiocParallel backend. SnowParam is the portable choice on Windows:

backend <- BiocParallel::SnowParam(
    workers = 2,
    type = "SOCK",
    stop.on.error = TRUE,
    progressbar = FALSE
)

rewiringValidation <- permuteRewiringScores(
    analysisData,
    candidateEdgeTable = knowledgeNetwork,
    groupColumn = "clinicalGroup",
    groupLevels = c("Recovered", "PASC"),
    minimumAbsoluteCorrelation = 0,
    adjustedPValueThreshold = NULL,
    nPermutations = 999,
    seed = 1,
    verbose = TRUE,
    progressEvery = 300,
    BPPARAM = backend
)

See vignette("CorNetto") for the full workflow.

Citation

If you use or build upon CorNetto, please cite it. The full author list, affiliations and a formatted reference are returned by:

citation("CorNetto")

A machine-readable citation is also available in CITATION.cff.

License

CorNetto is released under the Artistic License 2.0.