CLAUDE.md

In sbalci/ClinicoPathJamoviModule: Analysis for Clinicopathological Research

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Package Overview

ClinicoPath is a comprehensive jamovi module for clinicopathological research analysis. It provides statistical analysis tools specifically designed for pathology and clinical research, including survival analysis, decision analysis, descriptive statistics, and specialized plots.

Core Architecture

Jamovi Module Structure

This is a jamovi R module with a specific 4-file architecture pattern:

• .b.R files: Backend implementation classes (e.g., crosstable.b.R, survival.b.R)
• .a.yaml files: Analysis definitions with options/parameters (e.g., crosstable.a.yaml)
• .u.yaml files: User interface definitions (e.g., crosstable.u.yaml)
• .r.yaml files: Results/output definitions (e.g., crosstable.r.yaml)
• .h.R files: Auto-generated header files (compiled from .yaml files)

Key Backend Pattern

All analysis classes inherit from auto-generated base classes and use R6 class system:

crosstableClass <- R6::R6Class(
    "crosstableClass", 
    inherit = crosstableBase,  # Auto-generated from .yaml files
    private = list(
        .init = function() { ... },
        .run = function() { ... }
    )
)

Main Functional Areas

1. ClinicoPath Descriptives: Summary statistics, Table One, cross tables, data checking
2. ClinicoPath Survival: Survival analysis, Cox regression, Kaplan-Meier, competing risks
3. meddecide: Medical decision analysis, ROC curves, sensitivity/specificity, diagnostic tests
4. JJStatsPlot: Statistical plots using ggstatsplot wrappers
5. Patient Follow-Up Plots: Swimmer plots, waterfall plots, treatment response visualization

Development Commands

Testing

# Run all tests
Rscript -e "devtools::test()"

# Run specific test file
Rscript -e "testthat::test_file('tests/testthat/test-roc.R')"

Building and Checking

# Check package
Rscript -e "devtools::check()"

# Build package
Rscript -e "devtools::build()"

# Install development version
Rscript -e "devtools::install()"

Jamovi Module Development

# Build jamovi module (.jmo file)
# This requires jamovi development tools
Rscript -e "jmvtools::prepare()"
Rscript -e "devtools::document()"
Rscript -e "jmvtools::install()"

# Install module in jamovi for testing
# Copy .jmo file to jamovi modules directory

Key Dependencies and Patterns

Core Dependencies

• jmvcore: Jamovi module framework
• R6: Class system for all analysis classes
• magrittr: Pipe operators
• finalfit: Survival analysis and medical tables
• ggstatsplot: Statistical plotting
• gtsummary: Summary tables
• survival/survminer: Survival analysis

Data Processing Pattern

Most modules follow this pattern:

1. Clean variable names with janitor::clean_names()
2. Set variable labels with labelled::set_variable_labels()
3. Build formulas dynamically based on user inputs
4. Apply appropriate statistical tests
5. Format results for display

Output Types

• HTML tables: Primary output format for most analyses
• Plots: ggplot2-based visualizations
• Text summaries: Natural language interpretations
• Export options: CSV downloads for some modules

Module Categories by Menu Structure

Exploration Menu

• ClinicoPath Descriptives: tableone, summarydata, reportcat, benford
• ClinicoPath Comparisons: crosstable, chisqposttest
• ClinicoPath Descriptive Plots: agepyramid, alluvial, vartree, venn
• Patient Follow-Up Plots: waterfall, swimmerplot

Survival Menu

• ClinicoPath Survival: survival, survivalcont, multisurvival, oddsratio, singlearm

meddecide Menu

• Agreement: agreement, icccoeff
• Decision: decision, decisioncalculator, nogoldstandard, decisioncompare
• ROC: psychopdaroc, roc
• Power Analysis: kappasizeci, kappasizefixedn, kappasizepower

JJStatsPlot Menu

• Categorical vs Continuous: jjbetweenstats, jjwithinstats, jjdotplotstats
• Categorical vs Categorical: jjbarstats, jjpiestats
• Continuous vs Continuous: jjcorrmat, jjscatterstats

Common Development Tasks

Adding New Analysis Module

1. Create 4 jamovi files: .a.yaml, .u.yaml, .r.yaml, .b.R
2. Add entry to jamovi/0000.yaml
3. Implement R6 class inheriting from auto-generated base
4. Add to appropriate menu group/subgroup

Testing Strategy

• Unit tests in tests/testthat/
• Test data in data/ directory (many .csv, .omv, .rda files)
• Example analyses in jamovi format (.omv files)

Documentation

• Function documentation in .R files using roxygen2
• Vignettes in vignettes/ directory using R Markdown/Quarto
• Package website built with pkgdown

File Structure Notes

• R/: Backend R code (.b.R analysis classes, utility functions)
• jamovi/: Module definitions (.yaml files)
• data/: Example datasets in multiple formats
• man/: Auto-generated documentation
• inst/extdata/: Additional example data files
• vignettes/: Documentation and tutorials

sbalci/ClinicoPathJamoviModule documentation built on June 13, 2025, 9:34 a.m.