Supplementary data: Collection of publicly available MOF datasets and Random Forest Regressor ML Models trained with Database_07

This directory contains all datasets, machine learning (ML) models, and feature calculations associated with the research thesis “Accelerating the Discovery of Metal-Organic Frameworks for Carbon Capture: Efficient Machine Learning Predictions of CO₂ Uptake Based on Structural and Chemical Descriptors.” This thesis was conducted as part of a collaborative exchange program between the Department of Chemical and Biological Engineering (CHBE) at the University of British Columbia and the Institute of Energy Economics and Rational Energy Use (IER) at the University of Stuttgart. The structure of the directory generally follows the organization outlined in Table 2 of the thesis. All datasets are provided as ZIP archives. Each dataset folder contains the original CIF files and, where applicable, additional CSV files with calculated structural or chemical features and/or CO₂ uptake labels. In addition, the directory includes a folder “Calculations_Models_and_Plots” containing the trained ML models, Python code for data processing and feature calculation, and scripts used for training, evaluation, and plotting.