Enhanced validation of small-molecule ligands and carbohydrates in the Protein Data Bank

Abstract

The Worldwide Protein Data Bank (wwPDB) has provided validation reports based on recommendations from community Validation Task Forces for structures in the PDB since 2013. To further enhance validation of small molecules as recommended from the 2016 Ligand Validation Workshop, wwPDB, Global Phasing Ltd., and the Noguchi Institute, recently formed a public/private partnership to incorporate some of their software tools into the wwPDB validation package. Augmented wwPDB validation report features include: two-dimensional (2D) diagrams of small-molecule ligands and carbohydrates, highlighting geometric validation outcomes; 2D topological diagrams of oligosaccharides present in branched entities generated using 2D Symbol Nomenclature for Glycan representation; and views of 3D electron density maps for ligands and carbohydrates, illustrating the goodness-of-fit between the atomic structure and experimental data (X-ray crystallographic structures only). These improvements will impact confidence in ligand conformation and ligand-macromolecular interactions that will aid in understanding biochemical function and contribute to small-molecule drug discovery.

Keywords: 3D macromolecular structure; OneDep; PDB; Protein Data Bank; Worldwide Protein Data Bank; biocuration; small-molecule ligands; structural biology; structure data quality; validation.

Figure 1.

Cumulative growth of unique Chemical Components (ligands) in the PDB archive from 2000–2020.

Figure 2.

2D depictions of Mogul quality analysis of bond lengths, bond angles, torsion angles, and ring geometry, and 3D depictions of the atomic model fit to experimental electron density map (drawn in gray) with the surrounding difference map features (drawn in green and magenta) for NADP. Each fit is shown from a different orientation to approximate a three-dimensional view. (Left) PDB ID 1ZK4 represents a lower structure and/or data quality, with multiple features in the difference map. (Right) PDB ID 5ZIX represents higher structure and/or data quality, with a good fit of the model to the experimental electron density map and an essentially featureless difference map.

Figure 3.

Carbohydrate validation for PDB ID 1B5F. (A) 2D SNFG symbol image that indicates carbohydrate anomer (α versus β) and connectivity positions (1–3, 1–4 or 1–6 glycosidic connection) for oligosaccharide chain G. (B) 2D depiction of Mogul quality analysis of bond lengths, bond angles, torsion angles, and ring geometry (top), and three orthogonal projection views of the superposition of the atomic coordinates for oligosaccharide chain G onto the electron density map (in gray), with the very few significant features in the difference map in green and magenta (bottom).