ISCAPE

ISCAPE: Interpretable Support vector Classifier of Antibacterial activity of Peptides against Escherichia coli

Paper: Interpretable support vector classifier for reliable prediction of antibacterial activities of modified peptides against Escherichia coli

Dependencies:

Packages required to run the notebooks:

• Python 3.8
• pyforest 1.1.0
• umap-learn 0.5.6
• scikit-optimize 0.10.1
• xgboost 1.7.6
• lightgbm 4.0.0
• deepchem 2.7.1
• deep-forest 0.1.7
• shap 0.42.1

Colab notebooks:

• algo_search: contain the codes used in the model development
• models_pred: shows the predictions of all models on our own short peptide dataset
• ISCAPE: codes for ISCAPE model; data can be replaced with your own data to make peptide activity prediction and important feature identification

How to cite:

@article{Salas2026,
author = {Remmer L. Salas and Portia Mahal G. Sabido and Ricky B. Nellas},
title = {Interpretable support vector classifier for reliable prediction of antibacterial activity of modified peptides against Escherichia coli},
journal = {Journal of Molecular Graphics and Modelling},
volume = {142},
pages = {109188},
year = {2026},
issn = {1093-3263},
doi = {https://doi.org/10.1016/j.jmgm.2025.109188},
url = {https://www.sciencedirect.com/science/article/pii/S1093326325002487},
}