Welcome to the official documentation for HLP, an in silico method developed to predict the half-life of peptides in an intestine-like environment. While peptides are excellent drug candidates due to their specificity and low toxicity, their use is often hindered by high susceptibility to protease degradation. HLP allows researchers to design peptides with optimized physicochemical properties and improved stability.
Web Server: http://crdd.osdd.net/raghava/hlp/(https://webs.iiitd.edu.in/raghava/hlp)
Sharma, A., Singla, D., Rashid, M., & Raghava, G. P. S. (2014).
Designing of peptides with desired half-life in intestine-like environment. BMC Bioinformatics, 15, 282.
https://doi.org/10.1186/1471-2105-15-282
Zonedo:-(https://doi.org/10.5281/zenodo.20096828)
The HLP platform addresses the critical challenge of peptide stability. By utilizing Support Vector Machine (SVM) models trained on experimentally determined half-life data, HLP can predict how long a peptide will remain intact in a proteolytic environment.
The models were developed using two primary datasets of peptides with known half-lives in an intestine-like environment:
- HL10: A dataset of 10-mer peptides.
- HL16: A dataset of 16-mer peptides.
- Stability Prediction: Predicts the half-life of 10-mer and 16-mer peptides based on their amino acid, dipeptide, and tripeptide compositions.
- Physicochemical Profiling: Calculates essential properties such as hydrophobicity, charge, and molecular weight to understand their influence on peptide stability.
- Mutant Design: Enables the design of peptide analogs with enhanced stability by identifying specific residues that contribute to rapid degradation.
The SVM models achieved high correlation coefficients (R) between predicted and actual half-lives:
- HL10 Dataset: Maximum R of 0.69 using tripeptide composition.
- HL16 Dataset: Reliable performance across different composition-based models.
HLP utilizes a comprehensive range of sequence-based features to model the relationship between peptide structure and proteolytic resistance.
- Machine Learning: Developed using SVM-light and validated through rigorous cross-validation techniques.
- Input Features: Includes amino acid composition, dipeptide composition, and various physicochemical descriptors.
- Environment Simulation: The models specifically simulate the degradation patterns found in an intestine-like (protease-rich) environment.
- Drug Development: Optimizing the half-life of therapeutic peptides to ensure they reach their targets before being degraded.
- Oral Delivery Research: Assessing the feasibility of delivering peptides orally by predicting their survival in the gastrointestinal tract.
- Bioinformatics: Providing a reliable tool for large-scale screening and design of protease-resistant peptides.
Prof. Gajendra P. S. Raghava Bioinformatics Centre, CSIR-Institute of Microbial Technology, Chandigarh, India.
Email: raghava@imtech.res.in
This resource is open-access and distributed under the terms of the Creative Commons Attribution License, permitting unrestricted use and distribution provided the original work is properly credited.