This repository contains all the code and resources for the project titled "Applying an interpretable machine learning approach to assess intraspecific trait variation under landscape-scale population differentiation". The project focuses on the application of eXplainable Artificial Intelligence (XAI) techniques to assess intraspecific trait variations in ecology.
The research discusses functional traits in a dataset, emphasizing the missing values for each trait. The study uses visualizations to represent the percentage of missing values for each functional trait and the predictions made by a Random Forest (RF) classifier concerning populations of H. annuus. The data encompasses a wide range of traits reflecting plant architecture, reproductive phenology, tissue chemistry, and morphology of various plant parts. The combined dataset consists of 88 traits from 464 genotypes belonging to 49 populations.
- Processed_Datasets: 📊 Contains datasets that have been processed and are ready for analysis.
- Raw_Datasets_and_Tables: 📄 Contains the raw datasets and tables used in the research.
- Shape_files: 🌍 Contains shape files for geographical data visualization.
- 8_Dashboard.html: 🖥️ An interactive dashboard for data visualization.
- Applying-XAI-approaches-to-ecology.Rproj: 📈 R project file for the research.
- R and ArcGIS
Feel free to fork this repository, make changes, and create a pull request if you think you've made improvements worth sharing.
This research is under the bioRxiv license, and the DOI for the preprint is https://doi.org/10.1101/2023.04.07.536012, posted on April 8, 2023.