{
  "id": "skill_a99d7b18e7694005b651a299b906fd3e",
  "name": "bulk-rnaseq-differential-expression-multi-method",
  "description": "General-purpose bulk RNA-seq differential expression (DEG) analysis with automatic, data-driven method selection across DESeq2, edgeR (quasi-likelihood F-test), and limma-voom. Use this whenever someone has a bulk RNA-seq count matrix (genes x samples) and wants differentially expressed genes between conditions — treated vs control, KO vs WT, tumor vs normal, dose-response, time course — even if they don't name a method, and especially when they're unsure which of DESeq2/edgeR/limma-voom to use. The skill inspects the data (raw integer counts vs normalized, replicates per group, library-size spread, batch/paired structure, design rank), recommends the most appropriate method with plain-language reasoning, asks the user to confirm, then runs a unified workflow producing a standardized results table, QC plots (PCA, MA, volcano, mean-variance), significant-gene lists (BH-FDR padj), and an optional cross-method concordance check. Handles two-group and multi-group/contrast designs, batch and paired covariates, and tximport/featureCounts/SummarizedExperiment inputs.",
  "category": "general",
  "owner_type": "user",
  "created_by_name": "Tommer Schwarz",
  "verified": false,
  "icon": "flask",
  "keywords": [],
  "default_message": "",
  "created_at": "2026-06-03T00:15:37.850000",
  "updated_at": "2026-06-03T03:42:02.634000",
  "entrypoint": "SKILL.md",
  "supporting_files": {
    "references/comparison-and-caveats.md": {
      "type": "md",
      "can_edit": false,
      "editor_kind": null
    },
    "references/method-selection-guide.md": {
      "type": "md",
      "can_edit": false,
      "editor_kind": null
    },
    "references/troubleshooting.md": {
      "type": "md",
      "can_edit": false,
      "editor_kind": null
    },
    "scripts/concordance.R": {
      "type": "R",
      "can_edit": false,
      "editor_kind": null
    },
    "scripts/export_results.R": {
      "type": "R",
      "can_edit": false,
      "editor_kind": null
    },
    "scripts/filter_counts.R": {
      "type": "R",
      "can_edit": false,
      "editor_kind": null
    },
    "scripts/inspect_and_recommend.R": {
      "type": "R",
      "can_edit": false,
      "editor_kind": null
    },
    "scripts/load_data.R": {
      "type": "R",
      "can_edit": false,
      "editor_kind": null
    },
    "scripts/qc_plots.R": {
      "type": "R",
      "can_edit": false,
      "editor_kind": null
    },
    "scripts/run_deseq2.R": {
      "type": "R",
      "can_edit": false,
      "editor_kind": null
    },
    "scripts/run_edger.R": {
      "type": "R",
      "can_edit": false,
      "editor_kind": null
    },
    "scripts/run_limma_voom.R": {
      "type": "R",
      "can_edit": false,
      "editor_kind": null
    },
    "scripts/selftest.R": {
      "type": "R",
      "can_edit": false,
      "editor_kind": null
    }
  },
  "_category_display": "General"
}