A fully automatic evolutionary classification of protein folds: Dali Domain Dictionary version 3

Area of Science:

• Structural biology
• Bioinformatics
• Computational biology

Background:

• The Dali Domain Dictionary provides a numerical taxonomy of Protein Data Bank (PDB) structures.
• Existing classification relies on structural, functional, and sequence similarities.
• There was a need for a more comprehensive hierarchical classification.

Purpose of the Study:

• To extend the Dali classification to four traditional hierarchical levels.
• To computationally define supersecondary structural motifs (attractors) and remote homologues (functional families).
• To enhance the understanding of protein domain architecture and relationships.

Main Methods:

• Utilized measurements of structural, functional, and sequence similarities.
• Developed new computational definitions for attractors and functional families.
• Integrated data from the Protein Data Bank (PDB) and HSSP database.

Main Results:

• The Dali classification was extended to include supersecondary motifs, fold types, functional families, and sequence families.
• In September 2000, the database contained 10,531 PDB entries and 17,101 chains.
• Classified into five attractor regions, 1375 fold types, 2582 functional families, and 3724 domain sequence families.
• Associated sequence families with 99,582 unique homologous sequences from the HSSP database.

Conclusions:

• The extended Dali classification provides a detailed framework for protein domain architecture.
• It facilitates the identification of structural neighbors, conserved cores, and alignments of related protein families.
• This resource significantly increases the effective number of known protein structures and their relationships.