SpotOn

SpotOn identifies and classifies interfacial residues at protein-protein interfaces as Hot-Spots (HS) and Null-Spots (NS) using an ensemble machine learning approach.

Key Features:

• Hot-Spot and Null-Spot Classification: SpotOn classifies interfacial residues as Hot-Spots (HS) and Null-Spots (NS) with a demonstrated accuracy of 0.95 and sensitivity of 0.98 on an independent test set.
• Ensemble Machine Learning: The predictor uses an ensemble machine learning approach with up-sampling of the minor class.
• Structure and Sequence Features: SpotOn incorporates various features based on both protein 3D structure and sequence for prediction.

Scientific Applications:

• Protein-Protein Interface Analysis: SpotOn enables the identification of critical interfacial residues that contribute significantly to binding affinity at protein-protein interfaces.
• Hot-Spot Prediction for Protein Complexes: SpotOn was trained on 53 complexes and can be applied to predict Hot-Spots and Null-Spots in protein-protein interaction studies.

Methodology:

SpotOn uses an ensemble machine learning approach with up-sampling of the minor class, trained on 53 complexes using various features based on both protein 3D structure and sequence.