SCOP

SCOP classifies protein domains by evolutionary and structural relationships, providing a hierarchical framework for analysis of protein structure and evolution.

Key Features:

• Hierarchical Classification: SCOP classifies protein domains into species, protein, family, superfamily, fold, and class levels, with family and superfamily capturing near and distant evolutionary relationships and fold describing geometric relationships based on structural similarity.
• Evolutionary versus Physicochemical Distinction: Differentiates evolutionary relationships from similarities arising from the physics and chemistry of proteins.
• Comprehensive Data Integration: Integrates sequence data with structural information and provides links to atomic coordinates and literature references for each protein entry.
• Search Facilities: Provides a homology search that accepts input sequences to retrieve structurally similar proteins and a keyword search that matches database text and Brookhaven Protein Databank headers.
• Batch Classification and Updates: Employs batch classification protocols for new protein structures at the family and superfamily levels and incrementally refines classifications across releases.
• Integration with Other Databases: Participates in projects to integrate protein sequence and structure data across multiple databases to enable dynamic cross-database links.
• ASTRAL Sequence Libraries: Supplies sequences that form the basis of ASTRAL sequence libraries used to calibrate sequence search algorithms and generate statistics on protein structures.
• Access to Homologous Sequences (PDB-ISL): Via association with PDB-ISL enables matching of sequences from proteins with unknown structures to distantly related known structures using pairwise sequence comparison methods.

Scientific Applications:

• Protein Structure Prediction: Provides classified structural templates and evolutionary context to inform prediction of uncharacterized protein structures.
• Evolutionary Studies: Enables analysis of evolutionary pathways and relationships among protein families and superfamilies using its hierarchical classification.
• Drug Discovery and Design: Supports identification of structurally related targets and analysis of structural similarities and differences relevant to target identification and ligand design.

Methodology:

Combines manual inspection with automated methods, including batch classification protocols, homology searches and pairwise sequence comparison methods, and produces ASTRAL sequence libraries for algorithm calibration and structural statistics.