PRBP

PRBP predicts RNA-binding proteins (RBPs) from amino acid sequences by identifying RNA-binding residues and classifying proteins with a random forest that uses evolutionary information combined with physicochemical features (EIPP) and amino acid composition.

Key Features:

• Predictive Approach: Identifies RNA-binding residues within protein sequences and uses their presence or absence to inform protein-level RNA-binding prediction.
• Random Forest Methodology: Applies a random forest classifier when residue-based predictions are inconclusive, using evolutionary information combined with physicochemical features (EIPP) and amino acid composition as input features.
• Feature Analysis: Reports that EIPP substantially contributes to prediction accuracy and that incorporating RNA-binding residue predictions enhances overall performance.

Scientific Applications:

• Functional Annotation: Assists in annotating protein sequences by identifying potential RNA-binding capabilities.
• Protein Function Prediction: Supports prediction of protein functions related to RNA interactions in cellular processes.
• Drug Discovery and Development: Aids identification of proteins involved in RNA–protein interactions as candidate therapeutic targets.

Methodology:

Uses a two-step computational process: predict RNA-binding residues from sequence, and if residue-based results are insufficient, apply a random forest model using EIPP and amino acid composition features.