Urban 2015 Extinction Risk Analysis (WSL + R Pipeline)

Overview

This project implements a reproducible Linux (WSL) + R data analysis pipeline to explore extinction risk estimates from the Urban (2015) dataset from CSB. The workflow emphasizes data validation, lightweight preprocessing, and statistical analysis without requiring a database, making it portable and transparent.

The dataset contains study-level extinction estimates across taxa, regions, prediction years, and modeling assumptions.

Tools & Technologies

• WSL (Ubuntu) – execution environment

• Bash – pipeline orchestration and automation

• Python (standard library + pandas) – data profiling and preprocessing

• R – statistical analysis and visualization

• Git/GitHub – version control and reproducibility

Repository Structure

.
├── data/
│   ├── raw/              # Original dataset (TSV)
│   └── processed/        # Cleaned and grouped intermediate files
├── scripts/              # Bash pipeline scripts
├── R/                    # R analysis scripts
├── reports/              # Figures and analysis outputs
├── logs/                 # Pipeline logs
└── README.md

How to Run (from repo root in WSL)

chmod +x scripts/*.sh # make scripts executable

./scripts/run_all.sh # Run full pipeline end-to-end

Or run step-by-step.

./scripts/01_profile_validate.sh
./scripts/02_build_intermediates.sh
Rscript R/01_analysis.R

Pipeline Steps

1. Data profiling and validation (Bash + Python).

2. Intermediate data construction (Bash + Python).

3. Statistical analysis & visualization (R).

Key Outputs

• reports/overall_weighted.csv

• reports/fig_weighted_by_region_taxa.png

• reports/fig_threshold_sensitivity.png

• reports/percent_check_top25.csv (data quality check)

Reproducibility

All results can be regenerated on any Linux or WSL system with Bash, Python, and R installed by running the pipeline scripts provided in this repository.