Surface ID

Surface ID compares protein molecular surfaces using geometric deep learning to identify functional and interaction-relevant surface similarities.

Key Features:

• Geometric Deep Learning: Uses a geometric deep learning framework to analyze the complex geometry of protein molecular surfaces.
• Chemical Feature Integration: Integrates chemical attributes of surfaces alongside geometric features to improve functional discrimination.
• Surface and Structural Representations: Operates on representations including molecular surfaces, 3D atom coordinates, and 1D protein sequences.
• High-throughput Surface Comparison: Enables large-scale comparison of protein surfaces for systematic similarity assessments.
• Grouping and Alignment Algorithm: Implements a novel grouping and alignment algorithm to cluster and align surface regions based on similarity.

Scientific Applications:

• Protein Functional Annotation: Assesses surface similarity to support annotation of protein functions and identification of functionally related proteins.
• Visualization and In Silico Screening: Supports visualization of molecular surfaces and screening of potential binding partners for target molecules in drug discovery workflows.
• Clustering Proteins by Function: Groups proteins with similar surface features to facilitate identification of functionally related protein families and interaction partners.

Methodology:

Surface ID employs a geometric deep learning framework that integrates geometric and chemical features of molecular surfaces and applies a grouping and alignment algorithm for surface comparison and alignment.