Data and Code for Anticancer Peptide Prediction Using Machine Learning

Aegis is an end-to-end computational pipeline for identifying anticancer peptides (ACPs) from peptide sequences and visualising model performance. This record contains the source code and associated datasets used in the study, supporting reproducible machine learning–based ACP prediction. The pipeline covers data preprocessing, feature extraction, model training, probability export, and performance visualisation. Raw peptide files are cleaned and merged into unified feature matrices, followed by model training and evaluation. The provided scripts generate publication-ready figures, including ROC and precision–recall curves, incremental feature selection (IFS) curves, and UMAP visualisations. The workflow is implemented in Python (≥3.9) and is compatible with probability outputs generated by iLearnPlus (≥v0.1.4). Users can reproduce all analyses and figures reported in the associated manuscript, or apply the pipeline to new peptide datasets for ACP prediction and benchmarking.