Ensemble-AHTPpred

Ensemble-AHTPpred predicts antihypertensive peptides (AHTPs) using an ensemble machine learning approach to prioritize candidates for experimental validation.

Key Features:

• Ensemble Machine Learning Algorithms: Integrates Random Forest (RF), Support Vector Machine (SVM), and Extreme Gradient Boosting (XGB) in an ensemble to enhance prediction robustness and accuracy.
• Comprehensive Feature Set: Utilizes computed features including physicochemical properties, amino acid compositions (AACs), transitions, n-grams, and secondary structure-related information.
• Composite Feature Integration: Incorporates a composite feature generated through a logistic regression function that synthesizes multiple peptide characteristics.
• Performance Validation: Demonstrated overall accuracy exceeding 90% on independent test data.

Scientific Applications:

• Screening and Prioritization: Computationally screens potential antihypertensive peptide candidates prior to experimental validation.
• Resource Reduction: Reduces experimental time and resources by prioritizing high-confidence peptide candidates for laboratory testing.
• Prediction of Novel AHTPs: Capable of predicting experimentally validated AHTPs that were not part of the training or test datasets.

Methodology:

Integrates heterogeneous machine learning algorithms (RF, SVM, XGB) using a selected feature set of physicochemical properties, AACs, transitions, n-grams, and secondary structure-related information, incorporates a logistic regression–generated composite feature, and reports validation with overall accuracy exceeding 90% on independent test data.