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

Summary

Structures in PDB tend to contain errors. This is a very serious issue for authors that rely on such potentially problematic data. The community of structural biologists develops validation methods as countermeasures, which are also included in the PDB deposition system. But how are these validation efforts influencing the structure quality of subsequently published data? Which quality aspects are improving, and which remain problematic? We developed ValTrendsDB, a database that provides the results of an extensive exploratory analysis of relationships between quality criteria, size and metadata of biomacromolecules. Key input data are sourced from PDB. The discovered trends are presented via precomputed information-rich plots. ValTrendsDB also supports the visualization of a set of user-defined structures on top of general quality trends. Therefore, ValTrendsDB enables users to see the quality of structures published by selected author, laboratory or journal, discover quality outliers, etc. ValTrendsDB is updated weekly.

Availability and implementation

Freely accessible at http://ncbr.muni.cz/ValTrendsDB. The web interface was implemented in JavaScript. The database was implemented in C++.

Supplementary information

Supplementary data are available at Bioinformatics online.

Worldwide Protein Data Bank validation information: usage and trends

Realising the importance of assessing the quality of the biomolecular structures deposited in the Protein Data Bank (PDB), the Worldwide Protein Data Bank (wwPDB) partners established Validation Task Forces to obtain advice on the methods and standards to be used to validate structures determined by X-ray crystallography, nuclear magnetic resonance spectroscopy and three-dimensional electron cryo-microscopy. The resulting wwPDB validation pipeline is an integral part of the wwPDB OneDep deposition, biocuration and validation system. The wwPDB Validation Service webserver (https://validate.wwpdb.org) can be used to perform checks prior to deposition. Here, it is shown how validation metrics can be combined to produce an overall score that allows the ranking of macromolecular structures and domains in search results. The ValTrendsDB database provides users with a convenient way to access and analyse validation information and other properties of X-ray crystal structures in the PDB, including investigating trends in and correlations between different structure properties and validation metrics.

Validation of ligands in macromolecular structures determined by X-ray crystallography

Crystallographic studies of ligands bound to biological macromolecules (proteins and nucleic acids) play a crucial role in structure-guided drug discovery and design, and also provide atomic level insights into the physical chemistry of complex formation between macromolecules and ligands. The quality with which small-molecule ligands have been modelled in Protein Data Bank (PDB) entries has been, and continues to be, a matter of concern for many investigators. Correctly interpreting whether electron density found in a binding site is compatible with the soaked or co-crystallized ligand or represents water or buffer molecules is often far from trivial. The Worldwide PDB validation report (VR) provides a mechanism to highlight any major issues concerning the quality of the data and the model at the time of deposition and annotation, so the depositors can fix issues, resulting in improved data quality. The ligand-validation methods used in the generation of the current VRs are described in detail, including an examination of the metrics to assess both geometry and electron-density fit. It is found that the LLDF score currently used to identify ligand electron-density fit outliers can give misleading results and that better ligand-validation metrics are required.

How Does the Methodology of 3D Structure Preparation Influence the Quality of pKa Prediction?

The acid dissociation constant is an important molecular property, and it can be successfully predicted by Quantitative Structure-Property Relationship (QSPR) models, even for in silico designed molecules. We analyzed how the methodology of in silico 3D structure preparation influences the quality of QSPR models. Specifically, we evaluated and compared QSPR models based on six different 3D structure sources (DTP NCI, Pubchem, Balloon, Frog2, OpenBabel, and RDKit) combined with four different types of optimization. These analyses were performed for three classes of molecules (phenols, carboxylic acids, anilines), and the QSPR model descriptors were quantum mechanical (QM) and empirical partial atomic charges. Specifically, we developed 516 QSPR models and afterward systematically analyzed the influence of the 3D structure source and other factors on their quality. Our results confirmed that QSPR models based on partial atomic charges are able to predict pKa with high accuracy. We also confirmed that ab initio and semiempirical QM charges provide very accurate QSPR models and using empirical charges based on electronegativity equalization is also acceptable, as well as advantageous, because their calculation is very fast. On the other hand, Gasteiger-Marsili empirical charges are not applicable for pKa prediction. We later found that QSPR models for some classes of molecules (carboxylic acids) are less accurate. In this context, we compared the influence of different 3D structure sources. We found that an appropriate selection of 3D structure source and optimization method is essential for the successful QSPR modeling of pKa. Specifically, the 3D structures from the DTP NCI and Pubchem databases performed the best, as they provided very accurate QSPR models for all the tested molecular classes and charge calculation approaches, and they do not require optimization. Also, Frog2 performed very well. Other 3D structure sources can also be used but are not so robust, and an unfortunate combination of molecular class and charge calculation approach can produce weak QSPR models. Additionally, these 3D structures generally need optimization in order to produce good quality QSPR models.