A framework for protein structure classification and identification of novel protein structures

Area of Science:

• Computational biology
• Structural bioinformatics
• Protein science

Background:

• Protein structure classification is crucial for understanding protein function.
• The growing number of protein structures necessitates automated classification methods.

Purpose of the Study:

• To present a unified framework for automated protein structure classification and identification of novel protein structures.
• To accurately classify proteins into known folds and detect new protein folds.

Main Methods:

• Developed a unified framework comprising components for comparing, classifying, and clustering protein structures.
• Utilized the SCOP 1.69 database for evaluation.

Main Results:

• The proCC method achieved high accuracy in classifying new domains: 86.0% at family, 87.7% at superfamily, and 90.5% at fold levels.
• Successfully clustered novel protein domains, closely corresponding to new families in SCOP 1.69.
• Demonstrated the ability to suggest new classification groups containing novel folds.

Conclusions:

• The proCC method provides an effective and efficient solution for automatic protein domain classification and clustering.
• The method aids in identifying novel protein folds and suggesting new classification groups.
• A web-accessible version of proCC is available for public use.