Folddisco

Folddisco detects discontinuous or segmental three-dimensional (3D) protein motifs in large structural databases to identify functionally significant short 3D patterns.

Key Features:

• Position-Independent Geometric Indexing: Uses an index based on position-independent geometric features, including side-chain orientation, to represent discontinuous 3D motifs.
• Rarity-Based Scoring System: Implements a rarity-based scoring system to prioritize less common structural patterns that may be functionally significant.
• Efficiency and Scalability: Indexes 53 million AFDB50 structures into a compact 1.45 terabyte database within 24 hours, achieving roughly an order of magnitude speed improvement over existing methods.
• Accuracy and Storage Efficiency: Reports superior accuracy in detecting discontinuous motifs while being more storage-efficient compared with other available tools.

Scientific Applications:

• Motif Discovery: Identification of functionally crucial short 3D patterns within protein structures.
• Protein Function and Interaction Analysis: Facilitates understanding of protein function and interaction in structural biology research.
• Large-Scale Structural Database Mining: Enables motif discovery across large databases such as AFDB50.

Methodology:

Constructs a position-independent geometric index incorporating side-chain orientation, applies a rarity-based scoring system to rank motifs, and indexes AFDB50 structures (53 million) into a compact 1.45 terabyte database within 24 hours.