PFP-FunDSeqE

PFP-FunDSeqE predicts protein fold patterns by integrating protein functional domain annotations and evolutionary sequence data using a fusion ensemble classifier to classify proteins among 27 distinct fold patterns.

Key Features:

• Integration of Functional and Evolutionary Data: Combines protein functional domain annotations and evolutionary sequence data to leverage complementary biological information for fold prediction.
• Fusion Ensemble Classifier: Employs a fusion ensemble classifier that aggregates multiple models to improve prediction robustness and handle protein fold pattern complexity.
• Enhanced Predictive Success Rate: Demonstrated a success rate exceeding 70% in identifying proteins among 27 distinct fold patterns on a stringent benchmark dataset.

Scientific Applications:

• Protein structure–function analysis: Predicts fold patterns to support interpretation of protein functional conformations.
• Evolutionary biology and comparative genomics: Uses evolutionary sequence data to inform studies of protein evolutionary history and fold conservation.
• Structural genomics and drug design: Provides fold predictions that can guide target selection and therapeutic molecule development based on target protein folds.

Methodology:

Collects datasets containing functional domain annotations and evolutionary sequence data for proteins and processes these datasets through the fusion ensemble classifier to predict fold patterns.