This repository consists of a simple script with Python code to reproduce QSAR models presented in the publication:

"Hybrid machine learning and experimental studies of antiviral potential of ionic liquids against P100, MS2, and Phi6" by Szymon Zdybel, Anita Sosnowska, Dominika Kowalska, Julia Sommer, Beate Conrady, Patrick Mester, Maciej Gromelski and Tomasz Puzyn

Contents:

• ILS-SETS.xlsx - contains 3 sheets with data necessary to reproduce the models
• script.ipynb - simple Jupyter notebook script with Python code to reproduce the models, the script loads provided data from the xlsx file, fits models, performs predictions for the training/validation sets and generates classification report. These generated results are sufficient to reproduce any plot, graph and remaining statistics presented in the publication.
• LICENSE - self explanatory
• requirements.txt - lists all libraries and their versions that were used to develop QSAR models