Environmental niche modeling of ebolavirus occurrence in Uganda using MaxEnt, incorporating climate, vegetation, elevation, and human settlement predictors to identify high‐risk spillover zones.
# Ebola virus MaxEnt Ecological Niche Modeling (Uganda)
Reproducible pipeline, data, and interactive visualization for
**“Environmental niche modeling of ebolavirus occurrence in Uganda.”**
---
## Table of Contents
1. Overview
2. Repository Structure
3. Data Description
4. Installation & Environment Setup
5. Workflow 1: Data Preprocessing & Predictor Preparation
6. Workflow 2: MaxEnt Modeling & Analysis
7. Running the Pipeline
8. Interpreting & Visualizing Results
9. Reproducibility & Session Info
10. How to Cite
11. License
12. Zenodo Deposition
13. Contact Information
---
## Overview
This repository contains all data, code, and documentation needed to perform environmental niche modeling (ENM) of Sudan ebolavirus (SUDV) occurrence in Uganda. Using MaxEnt (Phillips et al., 2017), we identify areas of high ecological suitability for SUDV spillover based on climatic, topographic, vegetation, and human settlement predictors.
**Key Objectives:**
- Compile and clean georeferenced SUDV occurrence points (2000–2022).
- Assemble and preprocess high-resolution predictor rasters (bioclimatic variables, DEM, NDVI, rural settlement density).
- Implement a fully documented MaxEnt modeling pipeline in R (`maxent_pipeline.R`).
- Evaluate model performance using AUC, ROC curves, and omission rates.
- Generate both continuous and binary habitat suitability maps.
- Ensure full reproducibility via detailed workflows and session metadata.
---
## Repository Structure
```
sudv-enm-uganda/
├── README.md ← This file (project overview & instructions)
├── LICENSE ← CC BY 4.0 license text
├── CITATION.cff ← Citation metadata for GitHub/Zenodo
├── requirements.txt ← R package dependencies
├── zenodo.json ← Metadata for Zenodo deposition
├── data/
│ ├── occurrence/
│ │ └── sudv\_occurrence.csv ← Georeferenced SUDV spill …