Related Experiment Video
Updated: Jan 18, 2026
10:12
Construction of Cyclic Cell-Penetrating Peptides for Enhanced Penetration of Biological Barriers
Published on: September 19, 2022
2.7K
LightCPPgen: An explainable machine learning pipeline for rational design of cell penetrating peptides
Gabriele Maroni1, Filip Stojceski1, Lorenzo Pallante2
1Dalle Molle Institute for Artificial Intelligence IDSIA. USI/SUPSI, Lugano-Viganello, Switzerland.
International Journal of Antimicrobial Agents
|September 10, 2025
Summary
We developed LightCPPgen, a machine learning tool to design cell-penetrating peptides (CPPs) for drug delivery. This method accelerates the discovery of effective CPPs by reducing experimental time and costs.
Area of Science:
- Biotechnology
- Computational Biology
- Drug Delivery
Background:
- Cell-penetrating peptides (CPPs) are crucial for intracellular delivery of therapeutics.
- Rational design of CPPs is complex, requiring extensive experimental validation.
Purpose of the Study:
- To introduce LightCPPgen, an innovative de novo design strategy for CPPs.
- To leverage machine learning and optimization algorithms for efficient CPP sequence generation.
Main Methods:
- Developed a LightGBM-based predictive model using 20 explainable features for CPP translocation.
- Integrated the predictive model with a genetic algorithm (GA) for sequence optimization.
- GA prioritized penetrability while maintaining similarity to original peptide properties.
Main Results:
- Achieved accurate, efficient, and interpretable CPP prediction.
- Optimized CPP sequences for enhanced penetrability and retained biological properties.
- Significantly reduced time and cost for identifying promising CPP candidates.
Conclusions:
- LightCPPgen offers a robust framework for rational CPP design.
- The approach combines ML and optimization for explainable and interpretable peptide design.
- Facilitates accelerated development of CPP-based therapeutic delivery systems.
