Skip to content

STRUCTURE.md

Latest commit

 

History

History
65 lines (52 loc) · 2.33 KB

STRUCTURE.md

File metadata and controls

65 lines (52 loc) · 2.33 KB

Project Structure

This document describes the organization of the catecholamine research repository.

Directory Structure

catecholamine/
├── data/               # All data files
│   ├── raw/           # Immutable raw data files with metadata
│   └── derived/       # Generated datasets (rebuildable from raw)
├── figures/           # Output figures (generated, not committed)
├── notebooks/         # Exploratory analysis notebooks
├── paper/             # Paper manuscript and related files
│   └── sections/      # Paper sections (if split)
├── protocol/          # Research protocols and methodologies
├── refs/              # Bibliography and citation files
├── reports/           # Generated reports from scripts
├── schemas/           # Data validation schemas
├── scripts/           # Pipeline scripts
│   ├── scrape/       # Data scraping utilities
│   ├── validate_raw.py    # Validate raw data
│   ├── build_dataset.py   # Build master dataset
│   └── build_reports.py   # Generate reports
└── src/               # Reusable Python modules
    └── catecholamine/ # Main package

Key Files

  • README.md - Project overview and quickstart
  • requirements.txt - Python dependencies
  • pyproject.toml - Python package configuration
  • LICENSE - MIT license
  • CONTRIBUTING.md - Contribution guidelines
  • .gitignore - Git ignore patterns

Data Flow

  1. Raw Data (data/raw/) - Immutable source data with metadata
  2. Validation (scripts/validate_raw.py) - Ensure data quality
  3. Master Dataset (scripts/build_dataset.py) - Build normalized dataset
  4. Reports (scripts/build_reports.py) - Generate analysis reports
  5. Paper (paper/paper.qmd) - Render final manuscript

Important Principles

Data Provenance

  • Every raw data file must have a .meta.yaml companion file
  • Never edit raw data in place
  • All derived data must be rebuildable from raw data

Reproducibility

  • All figures generated from code
  • Paper rendered from Quarto documents
  • Complete dependency specification

Anti-Cherry-Picking

  • Document all inclusion/exclusion decisions
  • Log extraction methods
  • Use quality flags for questionable data

Getting Started

See README.md for setup instructions and workflow overview.