ECOD: new developments in the evolutionary classification of domains.

Area of Science:

• Structural biology
• Bioinformatics
• Evolutionary biology

Background:

• The Protein Data Bank (PDB) contains a vast number of protein structures.
• Classifying protein domains evolutionarily is crucial for understanding protein function and evolution.
• Existing classification systems may have limitations in scope or update frequency.

Purpose of the Study:

• To describe the Evolutionary Classification Of protein Domains (ECOD) database and its update process.
• To present the performance and improvements of the ECOD classification pipeline.
• To highlight new classifications and discuss future directions for ECOD.

Main Methods:

• ECOD utilizes a combination of automated and manual methods for weekly updates.
• Classification is based on known spatial structures from the Protein Data Bank (PDB).
• The system categorizes approximately 120,000 PDB depositions into over 500,000 domains within ~3,400 homologous groups.

Main Results:

• The ECOD weekly update pipeline demonstrates consistent performance.
• Improvements have been made to the ECOD website, enhancing search functionalities.
• Recent updates include the classification of novel protein structures and the identification of new homologous groups.

Conclusions:

• ECOD provides a comprehensive and up-to-date classification of protein domains.
• The database facilitates research in structural biology and evolutionary analysis.
• Future work will focus on refining the ECOD hierarchy and update procedures.