StackIL10: A stacking ensemble model for the improved prediction of IL-10 inducing peptides.

Area of Science:

• Immunology
• Bioinformatics
• Computational Biology

Background:

• Interleukin-10 (IL-10) is a cytokine with crucial anti-inflammatory roles in immunity.
• IL-10 can exhibit pro-inflammatory properties, necessitating identification of its inducing peptides.
• Manual identification methods are costly and inefficient.

Purpose of the Study:

• To develop a precise and efficient computational model for identifying IL-10-inducing peptides.
• To overcome the limitations of manual identification techniques.
• To leverage ensemble learning for improved peptide prediction.

Main Methods:

• Developed StackIL10, an ensemble learning model using stacking.
• Employed ten Amino-acid-composition-based Feature Extraction approaches.
• Utilized five Machine Learning Algorithms (LGBM, RF, SVM, Decision Tree, KNN) as base learners with Logistic Regression as the meta learner.

Main Results:

• StackIL10 achieved a 91.7% identification rate.
• The model demonstrated a Matthew Correlation Coefficient (MCC) of 0.833 and Specificity of 0.9078.
• StackIL10 outperformed single models and other ensemble combinations in identifying IL-10-inducing peptides.

Conclusions:

• StackIL10 offers a highly effective and efficient computational approach for identifying IL-10-inducing peptides.
• The ensemble learning strategy significantly enhances prediction accuracy.
• This model provides a valuable tool for immunological research and drug discovery.