MoRF-FUNCpred

MoRF-FUNCpred predicts functions of molecular recognition features (MoRFs) within intrinsically disordered regions (IDRs) of proteins to characterize MoRF-mediated interactions and functional roles.

Key Features:

• Multi-Label Learning Approach: Employs a multi-label learning framework to assign multiple functional labels to individual MoRFs.
• Binary Relevance Strategy: Uses the Binary Relevance (BR) strategy to decompose the multi-label task into independent binary classification problems.
• Ensemble Learning Techniques: Integrates Support Vector Machine (SVM), Logistic Regression (LR), Decision Tree (DT), and Random Forest (RF) classifiers into an ensemble to enhance prediction accuracy and robustness.
• Function Classification: Classifies MoRFs into five functional types: Molecular Recognition Assembler (MoR_assembler), Molecular Recognition Chaperone (MoR_chaperone), Molecular Recognition Display Sites (MoR_display_sites), Molecular Recognition Effector (MoR_effector), and Molecular Recognition Scavenger (MoR_scavenger).

Scientific Applications:

• Protein function annotation: Annotates functions of MoRF-containing IDRs to support interpretation of protein interaction networks.
• Disease mechanism and therapeutic development: Supports investigation of disease pathogenesis and pharmaceutical development by revealing MoRF-mediated interaction and functional sites relevant for drug design.

Methodology:

Treats MoRF function prediction as a multi-label learning task using Binary Relevance to create binary classifiers trained with SVM, LR, DT, and RF and combined via ensemble learning.