Homology-based loop modeling yields more complete crystallographic protein structures.

Area of Science:

• Structural biology
• Computational biology
• Biochemistry

Background:

• Protein structure determination via X-ray crystallography can be hindered by inherent flexibility, low-resolution data, or poorly defined electron density, leading to incomplete models.
• Advances in crystallographic refinement and model building techniques offer potential for completing previously missing structural regions.

Purpose of the Study:

• To develop and present algorithms for automatically identifying and incorporating missing protein regions from homologous structures into incomplete models.
• To refine and validate these newly incorporated regions through an automated procedure.

Main Methods:

• Algorithms were developed to identify missing regions in a target protein model by comparing it with homologous structures available in the Protein Data Bank (PDB).
• Identified missing regions were 'grafted' into the incomplete model and subsequently refined and validated using an automated pipeline.
• The developed methods were integrated into the PDB-REDO pipeline.

Main Results:

• The automated procedure successfully identified and incorporated 24,962 missing loops across various protein structures in the PDB.
• The PDB-REDO databank and webserver now provide access to these enhanced models and the automated procedures.
• The enhanced models contribute to a higher quality public archive of protein structures.

Conclusions:

• Automated methods for completing protein structural models significantly improve the completeness and quality of the Protein Data Bank.
• More complete protein structures facilitate a deeper understanding of protein function, enable more accurate comparisons between homologous proteins, and enhance structural bioinformatics data mining.