Skip to content

CLAUDE.md

Latest commit

 

History

History
150 lines (109 loc) · 3.48 KB

CLAUDE.md

File metadata and controls

150 lines (109 loc) · 3.48 KB

clover

R package for analyzing and visualizing nanopore tRNA sequencing data.

Session Workflow (READ FIRST)

At session start:

  1. Review R/ directory to understand existing patterns before adding code

Before creating new functions:

  1. Check existing files in R/ for similar functionality
  2. Follow established patterns (see Code Style section below)
  3. Add to existing files when appropriate rather than creating new ones

Before committing:

  • devtools::document() - regenerate documentation
  • devtools::test() - all tests pass
  • devtools::check() - no errors or warnings
  • New exported functions have roxygen2 documentation
  • NAMESPACE updated if exports changed

At session end:

  1. Commit all changes with descriptive message

File size limits:

  • No files > 5MB to git
  • Small sample data (.pod5, .bam, etc.) OK in inst/extdata/ if < 1MB
  • Large test data should be hosted externally and downloaded

Project Overview

Repository: https://github.com/rnabioco/clover Organization: RNA Bioscience Initiative, University of Colorado Anschutz Status: Active development (v0.0.0.9000)

Purpose

clover provides tools for:

  • Differential expression analysis of tRNAs
  • Base-calling error rate analysis and visualization
  • tRNA modification detection and comparison
  • Visualization including heatmaps and secondary structure diagrams

Data Model

The package uses SummarizedExperiment::RangedSummarizedExperiment to store:

  • counts: Expression levels per tRNA
  • bcerror: Base-calling error rates per tRNA and position
  • Reference sequences (FASTA)
  • Modification annotations

Development

Build System

  • Documentation: roxygen2 with markdown support
  • Testing: testthat (edition 3)
  • Vignettes: Quarto and knitr
  • Website: pkgdown with rbitemplate
  • Formatting: Posit AIR

Key Commands

# Install dependencies
pak::pak("rnabioco/clover")

# Build documentation
devtools::document()

# Run tests
devtools::test()

# Check package
devtools::check()

# Build pkgdown site
pkgdown::build_site()

Branch Strategy

  • devel: Main development branch (CI target)
  • Feature branches for new work

Code Style

R Conventions

  • Use tidyverse style (dplyr, tidyr, ggplot2)
  • roxygen2 for all exported functions
  • S4 classes extend Bioconductor base classes
  • Pipe-friendly function design

Documentation

  • All exported functions require roxygen2 documentation
  • Include @param, @return, @export, and @examples tags
  • Use markdown in roxygen comments (Roxygen: list(markdown = TRUE))

Function Patterns

# Exported function with roxygen2
#' Brief description
#'
#' @param arg description
#' @return description
#' @export
#' @examples
#' example_code()
function_name <- function(arg) {
  # implementation
}

Dependencies

Bioconductor

  • SummarizedExperiment
  • Biostrings

Tidyverse

  • dplyr, tidyr, readr, ggplot2, forcats

Testing

Tests live in tests/testthat/. Run with:

devtools::test()
testthat::test_file("tests/testthat/test-specific.R")

File Structure

R/                  # Source code
man/                # Generated documentation (do not edit)
tests/testthat/     # Unit tests
vignettes/          # Quarto vignettes
inst/extdata/       # Example data files
pkgdown/            # Website configuration

CI/CD

GitHub Actions workflows:

  • R-CMD-check.yaml: Package checks on Ubuntu (R devel + release)
  • pkgdown.yaml: Documentation site deployment