WESA

WESA predicts residue solvent accessibility in protein sequences, classifying residues as buried or exposed to support structural and functional analyses including hydrophobic core identification.

Key Features:

• Meta-predictor Approach: Employs a weighted ensemble/meta-predictor by averaging five distinct methods to classify residues into two states (buried or exposed).
• Hydrophobic Core Identification: Identifies residues that form the hydrophobic core by predicting buried residues relevant to protein stability.
• Predictive Accuracy: Reports an expected accuracy of 80% for two-state solvent accessibility prediction.

Scientific Applications:

• Deleterious Mutation Prediction: Aids identification of potential sites of deleterious mutations by flagging buried residues that are often mutation-sensitive.
• Structure Prediction: Provides solvent accessibility information that can be used to inform and improve protein structure prediction models.

Methodology:

Refines a suite of five prediction methods over an extensive dataset to produce a weighted ensemble average for two-state (buried/exposed) residue classification, with validation reported on six proteins with known mutation sites.