SCOPe

SCOPe classifies protein structures into a hierarchical framework that reflects their structural and evolutionary relationships to support structural analysis and variant interpretation.

Key Features:

• Comprehensive Classification: Classifies domains from proteins with known structures into a hierarchical framework using manual curation and automated methods.
• Integration with PDB Entries: Includes Protein Data Bank (PDB) entries; SCOPe 2.08-stable classifies 106,976 entries, approximately 60% of PDB entries.
• Machine-Parseable Annotations: Provides machine-parseable annotations that highlight aberrant structures and domains with smaller repeat units to support automated analyses and multiple sequence alignments of homologous proteins.
• Pfam Family Classification: Expands classification to include structures from 74 of the largest Pfam families not previously classified in SCOPe.
• Algorithmic Improvements: Refines algorithms to remove N- and C-terminal cloning, expression, and purification sequences from domain classifications.
• Backward Compatibility and Updates: Maintains backward compatibility with SCOP by providing consistent parseable files and a detailed history of changes between releases, and incorporates updates from the ASTRAL database.

Scientific Applications:

• Variant Interpretation: Maps genetic variants to classified protein structures to assess potential impacts on protein function and stability.
• Machine Learning Integration: Supplies machine-parseable annotations and comprehensive classifications for training deep learning classifiers and other automated analytical tools.
• Structural Biology Research: Supports studies of structural and evolutionary relationships among proteins, including analyses of protein function, interaction, and evolution.

Methodology:

Combines manual curation with automated methods to classify domains into a hierarchical framework, generates machine-parseable annotations highlighting aberrant and repeat-unit domains, applies algorithmic filtering to remove N- and C-terminal cloning/expression/purification sequences from domain definitions, and incorporates updates from the ASTRAL database.