The protein common assembly database (ProtCAD)-a comprehensive structural resource of protein complexes

Area of Science:

• Structural Biology
• Bioinformatics
• Computational Biology

Background:

• Protein assemblies are crucial for biological function, but their determination from experimental data can be ambiguous.
• Existing Protein Data Bank (PDB) annotations for biological assemblies may be incorrect or ambiguous in an estimated 10-15% of entries.
• Accurate protein assembly structures are vital for interpreting functional data and developing predictive deep learning models.

Purpose of the Study:

• To introduce the Protein Common Assembly Database (ProtCAD) for organizing and validating protein assembly structures.
• To provide a resource for clustering protein assembly structures derived from independent experimental determinations of homologous proteins.
• To enhance the reliability of biological assembly annotations in the PDB.

Main Methods:

• Development of the Protein Common Assembly Database (ProtCAD).
• Clustering of protein assembly structures from independent structure determinations within the Protein Data Bank (PDB).
• Integration of search functionalities (PDB entry, UniProt, Pfam) and download options (coordinates, PyMol scripts, annotations).

Main Results:

• ProtCAD clusters protein assembly structures from independent experiments, offering validated assemblies.
• Approximately 60% of PDB entries feature assemblies supported by at least two independent experiments.
• The database provides downloadable resources and publicly available assembly annotations for each cluster.

Conclusions:

• ProtCAD serves as a valuable resource for validating and accessing reliable protein assembly structures.
• The database aids in overcoming ambiguities in biological assembly annotations, improving data quality.
• ProtCAD facilitates downstream applications in functional genomics and the development of AI-driven structural biology tools.