run-r-analysis

Run R Analysis

A specialized sub-skill for executing R analyses in existing research projects and generating comprehensive outputs.

Overview

This sub-skill runs R scripts, RMarkdown documents, and Quarto reports within an existing R project structure. It captures all outputs including console results, plots, tables, and rendered reports.

Use this sub-skill when the user wants to:

• Execute an R script in a project
• Generate an R Markdown or Quarto report
• Run parameterized analyses with different inputs
• Capture and return analysis outputs

---

What This Sub-Skill Does

When invoked, this sub-skill will:

1. Activate project environment

   • Load renv environment if present
   • Set working directory to project root
   • Source .Rprofile if available

2. Execute analysis script

   • Run specified R script
   • Capture console output
   • Track generated files
   • Monitor for errors and warnings

3. Generate reports

   • Render R Markdown to HTML/PDF
   • Render Quarto documents
   • Include all figures and tables
   • Return rendered report path

4. Return results

   • Console output and messages
   • Paths to generated files
   • Summary statistics
   • Error messages if any

---

Example User Requests

• "Run analysis.R and show me the results"
• "Execute the differential expression analysis in my RNA-seq project"
• "Render the report and return the HTML file"
• "Run the analysis with different parameters"
• "Generate all plots from my analysis script"

---

Supported File Types

File Type	Execution Method	Output
.R scripts	source()	Console output + generated files
.Rmd	rmarkdown::render()	HTML/PDF/Word document
.qmd	quarto::quarto_render()	HTML/PDF/Word document
Parameterized scripts	with parameters	Multiple outputs

---

Execution Workflow

1. Verify project structure exists
   ↓
2. Activate renv environment (if present)
   ↓
3. Load required packages
   ↓
4. Execute analysis script
   ↓
5. Capture outputs
   ↓
6. Return results summary

---

Output Handling

Console Output

All print statements, messages, and warnings are captured and returned.

Generated Files

• Figures: results/figures/*.pdf, *.png, *.svg
• Tables: results/tables/*.csv, *.tsv
• Models: results/models/*.rds
• Reports: reports/*.html, *.pdf

Error Handling

• Script errors are reported with line numbers
• Missing packages are identified
• File path issues are diagnosed
• Suggestions for fixes are provided

---

Parameters

Parameter	Description	Default
script_path	Path to R script or Rmd file	Required
working_dir	Project root directory	Current directory
render_report	Also render Rmd/qmd if present	true
output_format	Report format (html, pdf, all)	html
params	Named list for parameterized reports	NULL

---

Example Usage

Basic Script Execution

User: Run the analysis script in my project.

Skill actions:

# Set working directory
setwd("my-project")

# Activate renv
renv::activate()

# Source analysis script
source("scripts/01_analysis.R")

# Report outputs
# - results/tables/statistics.csv created
# - results/figures/pca_plot.pdf created
# - results/models/regression_model.rds created

Parameterized Report

User: Run the report with dataset='variant2'.

Skill actions:

rmarkdown::render(
  "reports/report.Rmd",
  params = list(dataset = "variant2"),
  output_format = "html_document"
)
# Output: reports/report_variant2.html

Quarto Report

User: Render the Quarto document to PDF.

Skill actions:

quarto::quarto_render(
  "reports/analysis.qmd",
  output_format = "pdf"
)
# Output: reports/analysis.pdf

---

Common Output Patterns

Analysis Type	Expected Outputs
Differential Expression	Volcano plots, MA plots, DEG tables
PCA	PCA plots, loadings, explained variance
Clustering	Dendrograms, heatmaps, cluster assignments
Regression	Diagnostic plots, coefficient tables, predictions
Time Series	Forecast plots, residuals, accuracy metrics

---

Notes

• Working directory is always set to project root
• Relative paths in scripts work correctly
• renv ensures correct package versions
• All outputs are tracked and reported
• Long-running scripts show progress updates

---

Related Sub-Skills

• create-project: Set up new projects before running analyses
• debug-env: Troubleshoot environment issues before execution