BOM (Bag-of-Motifs) – Scripts for Manuscript Reproducibility

This release contains all scripts, workflows, and helper functions used to generate the analyses and figures in the BOM manuscript. The codebase is intended to ensure full reproducibility of the published results and to provide a foundation for extending the approach to new datasets.

Contents:
• Core scripts for BOM feature extraction, motif scanning, and sequence encoding.
• Training and evaluation pipelines for classifier models (including cross-validation and performance metrics).
• Figure generation scripts corresponding to the main and supplementary figures.
• Utility scripts for data preprocessing, normalization, and integration with external motif libraries.

Environment & Dependencies:
• Scripts are tested in R packages (R 4.0.0).
• Dependencies listed in requirements.txt.
• Example command lines provided in README.md.

Usage:
• Instructions for reproducing key results are in the README.md.
• Example datasets and config files are included for reference.
• Users can adapt the pipelines to their own motif or enhancer datasets.

Notes:
• Code is provided “as is” to support reproducibility of the manuscript results.
• For large-scale analyses, HPC/cluster execution may be required.
• Please cite the BOM manuscript when using or extending this code.