Changes in version 0.99.0                        

New features

  - Initial release submitted to Bioconductor.

  - loadHostData(): Import and normalize bulk RNA-seq count data using
    TMM normalization and limma-voom precision weights. Returns a
    SummarizedExperiment with both raw counts and voom-transformed
    values.

  - loadMicrobiomeData(): Import and normalize microbiome taxa tables
    using centered log-ratio (CLR) or total sum scaling (TSS)
    transformation, with automatic zero-inflation handling via
    pseudocounts. Returns a SummarizedExperiment.

  - matchSamples(): Identify and retain paired samples present in both
    host and microbiome datasets. Returns a MultiAssayExperiment.

  - jointDimReduction(): Run sparse multi-block PLS-DA (DIABLO from
    mixOmics) for joint dimensionality reduction across host and
    microbiome data layers. Returns integrated sample scores and feature
    loadings.

  - biomarkerDiscovery(): Identify top host genes and microbial taxa
    using DIABLO sparse loadings and cross-omics Spearman correlation
    networks. Returns a ranked DataFrame of multi-omics biomarkers.

  - diagnosticClassifier(): Train and evaluate host-only,
    microbiome-only, and joint Random Forest classifiers using nested
    cross-validation. Provides AUC-ROC comparison quantifying the
    multi-omics advantage.

  - MultiOmicsBridgeAnalysis(): One-call wrapper executing the full
    pipeline from matched MultiAssayExperiment through classification.

  - plotIntegration(): Joint biplot of sample scores in integrated
    component space with feature loading vectors.

  - plotBiomarkerNetwork(): Clustered heatmap of host gene to microbial
    taxon cross-omics Spearman correlations.

  - plotClassifierComparison(): ROC curves comparing host-only,
    microbiome-only, and joint classifier performance.

  - plotSankey(): Feature-flow diagram from omics layer through selected
    biomarkers to outcome classes.

  - generateReport(): Print or save a structured text summary of all
    analysis results.

  - MOBResult S4 class with slots for integrated scores, feature
    loadings, biomarker table, classifier performance, and parameters.
    Accessor methods: biomarkers(), performance(), integrationScores(),
    featureLoadings().