The Protein Data Bank (PDB) was established in 1971 as the first open-access digital data resource in biology with just seven X-ray structures of proteins. During its first 50 years of continuous operations, PDB holdings have grown to more than 175,000 structures becoming the single global archive of 3D-structures of proteins, nucleic acid, and their complexes with one another and small-molecule ligands. Open access to expertly biocurated PDB structures enables the efforts of many millions of basic and applied researchers, educators, and students around the world. Their work impacts fundamental biology, biomedicine, bioengineering, biotechnology, and energy sciences.

The Worldwide Protein Data Bank (wwPDB, wwpdb.org) manages the PDB archive according to the FACT principles of Fairness-Accuracy-Confidentiality-Transparency and the FAIR principles of Findable-Accessible-Interoperable-Reusable. Current wwPDB members include the US RCSB Protein Data Bank (RCSB PDB), Protein Data Bank in Europe (PDBe), Protein Data Bank Japan (PDBj), Electron Microscopy Data Bank (EMDB), and Biological Magnetic Resonance Bank (BMRB).

All data in the PDB archive conform to the wwPDB PDBx/mmCIF data dictionary, which is fully extensible both human- and machine-readable. PDB structures are composed of amino acids or nucleotide building blocks that comprise biopolymers, and associated small molecules such as water molecules, solute molecules, ions, co-factors, metabolites, enzyme inhibitors, drugs, etc. Every new structure coming into the PDB is processed using the wwPDB OneDep global system for deposition, validation, and biocuration. All PDB structures are accompanied by an official wwPDB Validation Report, exemplifying standards developed collaboratively with wwPDB Task Forces composed of community experts.

Small-molecule constituents of PDB structures are defined in the wwPDB Chemical Component Dictionary (CCD). This dictionary contains detailed chemical descriptions for standard and modified amino acids/nucleotides, small molecule ligands, solvent molecules, and others. Precise knowledge of interactions between macromolecules and small-molecule ligands is central to our understanding of biological and biochemical function, drug action, mechanisms of drug resistance, and drug-drug interactions.

Recent enhancements to the CCD and the wwPDB Validation Report will be described, together with value-added information concerning ligand quality now available on the US Research Collaboratory for Structural Bioinformatics Protein Data Bank PDB website (RCSB PDB, RCSB.org).

wwPDB members are US RCSB PDB (supported by NSF, NIH, DOE, and Rutgers Cancer Institute of New Jersey), PDBe (EMBL-EBI, Wellcome Trust, BBSRC, MRC, and EU), and PDBj (NBDC-JST), and BMRB (NIGMS).

Peer review of supporting data for submitted research articles is currently assuming great significance in scientific publishing, but is not new for the journals of the IUCr. Co-editors of Acta Crystallographica C under the editorship of Sidney Abrahams (1924-2021) were expected to validate the consistency of crystal structure data for reported structures, for which cell parameters and symmery, coordinates, geometry and anisotropic displacement parameters were mandatory. The journal developed software to reduce the calculational burden, and this evolved into the checkCIF service that allowed authors to participate in the validation effort, and to account for apparent anomalies or outliers in their derived structures. An early consequence was the almost complete elimination of corrigenda that resulted from post-publication surveys by individual scientists or by database aggregators. Over the years checkCIF increased in sophistication and power (authors were required to supply structure factors in machine-readable form), and has been adopted by other journal publishers and by structural databases. The IUCr Diffraction Data Deposition Working Group (2011-2017) emphasised the value of access to raw experimental data in evaluating structure interpretation, and IUCr Journals have responded by encouraging authors to make available their diffraction data sets. The journals continue to explore ways to improve the refereeing process with regard to data, in their effort to make the initial publication of the version of record of an article as error-free as possible.

The Cambridge Structural Database (CSD) was founded on a vision that collective use of data would lead to the discovery of new knowledge which transcends the results of individual experiments. Excellent data sharing practices in the crystallographic community as well as deposition and curation processes at the CCDC have enabled that vision to come to true.

This talk will demonstrate a number of ways in which a structural database such as the CSD can work with the community to help set standards from validation to publication and explore what part we play in the pre and post publication peer review of crystallographic data. We will share our experiences evolving our interactive deposition service, from the integration of checkCIF, to the establishment of links to raw diffraction data. We will also look at how repositories can support the peer review of data and what impact an increase of data published solely through the CSD might have. The presentation will conclude by looking at how reviewing crystallographic results might change in the future, how the CSD could evolve and how we can better help the community increase the integrity of data that is shared.

High-resolution macromolecular structures determined using crystallography, NMR, and cryo-EM provide a gold standard for evaluation of important properties of biomolecules, but the quality of some structures, as well of their presentation, is not always fully acceptable. Whereas quality checking tools provided by the PDB during deposition process may flag some common problems, the resulting red flags are not always addressed by deposition authors. Some journals require that manuscripts be accompanied by validation reports in order to assist reviewers in evaluation of the validity of presented structures, whereas other journals do not have such requirements. Additionally, validation reports are more helpful in identifying global problems, while some local problems may not be apparent. Utilization of additional tools and interactive software might assist readers in making the best use of published structural data. Using examples extracted from the Protein Data Bank, as well as from journal publications, some common problems will be identified and suggestions will be made on how to avoid their reoccurrence.

Over the last decades, the PDB has been developing tools and standards for the assessment of the quality of the structural models deposited in its archives. Similarly, more and more journals are now requiring validation reports generated by the PDB as a prerequisite for manuscript submission. Such quality metrics have been used previously to gauge the relationship between structural model quality and publication venues [1,2]. More recently, these indicators have been applied to assess the evolution of the quality of the PDB deposits with time [3] using the concept of a percentile (P_Q1) metric, which combines such measures as R_free, RSRZ (normalized Real Space R-factor) outliers, Ramachandran outliers, Rotamer outliers, and Clashscore.

In this paper we will show how the quality of macromolecular models deposited in the PDB has changed over the years (Fig. 1) and how the P_Q1 parameter can be converted to a new measure, P_Q1(t,d), that takes into account time (t) and data resolution (d). The proposed new measure can be used to reveal how structure quality in a given moment of time was related to such issues as:

• differences between proteins and nucleic acids;
• comparison with structural genomics projects;
• assessment of deposits without journal publications (To be published);
• journal impact factor (Fig. 2).

The paper will also discuss how the quality of crystallographic macromolecular structures in the PDB has improved over the last years and highlight some crucial periods in this history.

[1] Brown, E. N. & Ramaswamy, S. (2007). Acta Cryst. D63, 941–950.

[2] Read, R. J. & Kleywegt, G. J. (2009). Acta Cryst. D65, 140–147.

[3] Shao, C., Yang H., Westbrook, J. D., Young, J. Y., Zardecki, C. & Burley, S. K. (2017). Structure 25, 458–468.