ValTrendsDB: bringing Protein Data Bank validation information closer to the user

Area of Science:

• Structural Biology
• Bioinformatics
• Data Science

Background:

• The Protein Data Bank (PDB) is a critical resource for structural biology, but its deposited structures can contain errors.
• Validation methods are employed to assess and improve the quality of structural data.
• The impact of these validation efforts on the quality of subsequently published structures remains unclear.

Purpose of the Study:

• To investigate the influence of validation methods on the quality of biomacromolecular structures in the PDB.
• To identify which aspects of structure quality are improving and which remain problematic.
• To provide a tool for analyzing trends in structural data quality.

Main Methods:

• Development of ValTrendsDB, a database for analyzing relationships between quality criteria, size, and metadata of biomacromolecules.
• Utilizing data sourced from the PDB for extensive exploratory analysis.
• Presenting discovered trends through precomputed, information-rich plots.

Main Results:

• ValTrendsDB reveals trends in biomacromolecular structure quality over time.
• The database allows visualization of user-defined structures alongside general quality trends.
• Users can assess the quality of structures from specific authors, laboratories, or journals and identify outliers.

Conclusions:

• ValTrendsDB offers insights into the effectiveness of validation efforts in structural biology.
• The tool facilitates the identification of quality outliers and facilitates informed data selection.
• Continuous updates ensure ValTrendsDB reflects the latest trends in PDB structure quality.