A web server for predicting and scanning of IL-5 inducing peptides using alignment-free and alignment-based method.

Area of Science:

• Immunology
• Computational Biology
• Bioinformatics

Background:

• Interleukin-5 (IL-5) is a key therapeutic target for eosinophil-mediated diseases.
• Predicting IL-5 inducing antigenic regions is crucial for therapeutic development.

Purpose of the Study:

• To develop a highly precise computational model for predicting IL-5 inducing antigenic regions in proteins.
• To create a user-friendly tool for researchers to identify potential IL-5 epitopes.

Main Methods:

• Utilized experimentally validated peptide data from IEDB for training and testing.
• Developed and compared alignment-based, alignment-free (machine learning), and hybrid models.
• Evaluated models using Area Under the Curve (AUC) and Matthews Correlation Coefficient (MCC).

Main Results:

• Alignment-free models, particularly a dipeptide-based Random Forest, showed improved performance (AUC 0.75).
• A hybrid model combining alignment-based and alignment-free methods achieved the highest performance (AUC 0.94, MCC 0.60).
• Identified specific residues (Ile, Asn, Tyr) enriched in IL-5 inducing peptides.

Conclusions:

• The developed hybrid model significantly enhances the prediction of IL-5 inducing regions.
• The 'IL5pred' web server and standalone package provide a valuable resource for immunologists and drug developers.
• This work facilitates the design of targeted therapies for IL-5 driven diseases.