Protein Data Bank and Related Structural Archives

  • The Protein Data Bank (PDB) and related structural archives are essential resources in bioinformatics for storing, accessing, analyzing, and sharing three-dimensional structures of biological macromolecules. These resources provide structural information about proteins, nucleic acids, molecular complexes, and their interactions with ligands, ions, and other molecules. Structural data help researchers understand how biological molecules fold, recognize binding partners, catalyze biochemical reactions, and perform their functions. When combined with sequence databases, protein family resources, pathway databases, and computational analysis tools, structural archives support research in molecular biology, structural bioinformatics, drug discovery, biotechnology, and evolutionary biology.
  • The Protein Data Bank is the principal international archive of experimentally determined three-dimensional structures of biological macromolecules. It was established in 1971 and has developed into a global resource for structural biology. The PDB contains structures determined using methods such as X-ray crystallography, nuclear magnetic resonance (NMR) spectroscopy, and cryo-electron microscopy (cryo-EM). Its records include proteins, DNA, RNA, protein–nucleic acid complexes, molecular assemblies, and complexes containing small molecules or other ligands. Structural information is deposited by researchers and made available to the scientific community through coordinated international infrastructure.
  • The Worldwide Protein Data Bank (wwPDB) manages the international PDB archive through its member organizations, which include the Research Collaboratory for Structural Bioinformatics Protein Data Bank (RCSB PDB), the Protein Data Bank in Europe (PDBe), and the Protein Data Bank Japan (PDBj). These organizations coordinate deposition, validation, archival standards, and distribution of structural data. Although each organization offers different search interfaces, visualization tools, and additional services, they provide access to the same core international archive. The Biological Magnetic Resonance Data Bank (BMRB) is a related resource specializing in biomolecular NMR data, including experimental measurements and restraints that can support structure determination and analysis.
  • Each PDB structure is assigned a unique four-character accession identifier, often called a PDB ID. This identifier provides a convenient way to retrieve the corresponding structural record from the archive. A structure record may contain atomic coordinates, polymer sequences, descriptions of the biological macromolecules, experimental methods, citations, ligands, and information about the structural assembly. It may also include details about mutations, engineered constructs, missing residues, post-translational modifications, and experimental conditions. Researchers should examine these details before using a structure as evidence for a protein’s biological properties.
  • PDB structures are available in several file formats, including PDB and macromolecular Crystallographic Information File (mmCIF). The legacy PDB format is widely recognized and remains useful for many software applications, but it has limitations in representing some large structures and complex metadata. The mmCIF format supports more extensive identifiers, structural information, and metadata, making it the preferred archival format for many modern workflows. Related formats, such as PDBx/mmCIF and BinaryCIF, support structured data exchange and efficient processing. Researchers should choose formats according to the requirements of their analysis software and retain the original record when reproducibility is important.
  • The atomic coordinates in a structure describe the three-dimensional positions of atoms within the modeled biological molecule. These coordinates allow researchers to investigate protein folding, secondary and tertiary structure, ligand-binding sites, catalytic residues, and interactions between molecular components. Structural records may also contain information about solvent molecules, metal ions, cofactors, and small-molecule ligands. However, an experimentally determined structure does not necessarily represent every region or conformational state of a protein. Flexible regions may be unresolved, some residues may be absent from the coordinate model, and a structure may represent a particular state selected by the experimental conditions.
  • The experimental method used to determine a structure influences how the resulting data should be interpreted. X-ray crystallography determines structural information from diffraction by crystallized samples and is widely used to study proteins and macromolecular complexes. NMR spectroscopy provides structural and dynamic information about molecules in solution, particularly for suitable molecular sizes and experimental conditions. Cryo-EM uses images of rapidly frozen specimens to reconstruct three-dimensional structures and has become especially important for large molecular assemblies. Each technique has distinct strengths and limitations, so researchers should consider the experimental method, resolution or other relevant quality measures, model completeness, and validation information when evaluating a structural record.
  • Structure validation is an important component of structural archive quality control. Deposited structures undergo checks that may examine geometry, stereochemistry, atomic contacts, sequence consistency, and agreement between the model and experimental observations. Validation reports can reveal unusual bond lengths or angles, steric clashes, poorly supported regions, and other potential problems. These indicators should be interpreted in the context of the experimental method and the biological system. A single validation score does not fully determine whether a structure is biologically meaningful, but validation information helps researchers assess its suitability for a particular application.
  • The PDB archive is closely connected to other biological databases. Structural records may link to UniProt protein accessions, NCBI sequence records, organism information, protein family classifications, literature references, and small-molecule databases. These cross-references help researchers connect a three-dimensional structure with the protein’s amino acid sequence, gene, known function, evolutionary relatives, and biological context. Such links are particularly valuable when a structure represents only a fragment of a protein or a modified version of a natural molecule.
  • The PDB also provides information about biological assemblies, which describe the molecular arrangements believed to represent functional complexes. A crystallographic unit cell may contain multiple copies of a molecule because of the crystal-packing arrangement, but not every contact between neighboring molecules represents a physiological interaction. Biological assembly annotations attempt to identify the relevant functional arrangement using experimental and computational evidence. Researchers should therefore distinguish the biological assembly from the asymmetric unit or crystal packing when studying oligomerization, protein–protein interfaces, or ligand binding.
  • The Protein Data Bank in Europe, known as PDBe, is one of the major wwPDB portals. It provides structural search and visualization services, programmatic access, validation information, and connections to related biological resources. PDBe also supports structural interpretation through annotations and links to protein sequence and functional databases. Researchers can use its services to locate structures by protein sequence, organism, ligand, experimental method, or structural features. Its APIs and downloadable datasets support automated structural analysis and integration with computational pipelines.
  • RCSB PDB provides another major portal for searching and exploring PDB structures. Its services include sequence-based searches, structure visualization, ligand information, molecular annotations, and links to related records. Researchers can explore structural similarities, compare macromolecular assemblies, investigate bound compounds, and retrieve coordinate files for downstream analyses. RCSB PDB also offers programmatic access to structural metadata and search services, making it useful for large-scale studies and automated data retrieval.
  • PDBj, the Protein Data Bank Japan, contributes to the international PDB archive and provides tools for structural data exploration and analysis. Its services support structural searches, visualization, and access to molecular information. Like the other wwPDB partners, PDBj provides access to the shared PDB archive while offering its own interfaces and additional computational resources. These portals can be used interchangeably for many basic retrieval tasks, although individual features and interface capabilities may differ.
  • The Biological Magnetic Resonance Data Bank is an important complementary archive for NMR-related biological data. It stores information such as chemical shifts, experimental restraints, and other data used in biomolecular NMR research. These data can help researchers investigate molecular structure, flexibility, and conformational changes. BMRB information may be linked to PDB structures when NMR data contribute to a deposited structural model. Access to the underlying experimental measurements can be especially valuable when interpreting structural uncertainty or comparing alternative conformations.
  • The Electron Microscopy Data Bank (EMDB) is a major archive for three-dimensional electron microscopy maps and related structural data. It complements the PDB by providing density maps and other volumetric data generated by electron microscopy, including cryo-EM. A PDB coordinate model may be associated with an EMDB map when a macromolecular structure has been fitted or built into experimental density. Examining both the coordinate model and the underlying map can help researchers assess whether the proposed atomic structure is supported by the experimental observations.
  • The Small Angle Scattering Biological Data Bank (SASBDB) stores experimental small-angle X-ray scattering and small-angle neutron scattering data for biological macromolecules and complexes. These methods provide information about molecular size, shape, and overall organization in solution. They generally do not produce atomic-resolution structures on their own, but their data can help validate structural models, investigate flexible proteins, and compare molecular conformations under different conditions. SASBDB is therefore useful when a protein’s solution behavior differs from what might be inferred from a static high-resolution structure.
  • Predicted structural resources complement experimental archives. The AlphaFold Protein Structure Database provides predicted structures for many proteins, accompanied by confidence information that helps users assess the reliability of different regions. Unlike the experimental PDB archive, these predictions are generated computationally and should not be described as experimentally determined structures. They can support domain analysis, structural comparisons, hypothesis generation, and the selection of proteins for further experiments. Confidence varies across proteins and regions, and a predicted structure does not independently establish biological activity, ligand binding, or the correct conformational state of a protein complex.
  • Other structural resources include ModelArchive, which has provided a repository for computationally modeled macromolecular structures, and specialized collections of protein structures and structural annotations. The status, scope, and submission policies of individual repositories can change over time, so researchers should verify current availability and documentation before selecting a resource for a new project. Structural classifications such as CATH and SCOP are also important related resources, although they classify protein domains and folds rather than functioning as primary archives of experimental coordinate data.
  • PDB and related archives support a wide range of structural bioinformatics applications. Researchers can use structural records to identify active sites, examine conserved residues, compare protein folds, investigate molecular recognition, analyze protein–protein interactions, and study ligand-binding pockets. Structural comparisons can reveal evolutionary relationships between proteins that have limited sequence similarity. In drug discovery, experimentally determined structures can guide the design and optimization of compounds that interact with a target protein. However, structural similarity and predicted binding poses must be supported by appropriate biochemical, biophysical, or cellular evidence before making strong claims about activity or therapeutic potential.
  • Structural archives are also valuable for protein annotation and comparative genomics. When an uncharacterized protein resembles a structurally characterized protein, the known structure may provide clues about the query protein’s fold, domain organization, and possible functional regions. Researchers can compare amino acid sequences, identify conserved domains, and map functional residues onto related structures. Nevertheless, functional annotation should not rely on structural resemblance alone because proteins with similar folds may catalyze different reactions or participate in distinct biological processes. Sequence evidence, evolutionary relationships, biochemical data, and experimental context should be considered together.
  • Programmatic retrieval is important for large-scale structural analyses. Researchers can use database APIs, search services, downloadable coordinate files, and bulk data resources to automate the retrieval of structures matching particular proteins, organisms, ligands, or experimental methods. Automated pipelines may retrieve PDB identifiers, download mmCIF files, extract sequences, calculate structural descriptors, and compare structures using appropriate software. Such workflows should handle missing records, obsolete identifiers, alternate conformations, modified residues, and multiple structural chains. They should also respect service guidelines and document the source and version of the data used.
  • Structural archive identifiers and metadata are essential for reproducibility. A protein may have multiple PDB structures representing different conformations, ligand-bound states, mutants, domains, or experimental constructs. Researchers should record the PDB ID, relevant chain identifier, biological assembly, experimental method, and any structure version or revision information available. For predicted models, the database release and model identifier should also be retained. These details allow other researchers to identify the precise structural evidence used in an analysis and distinguish between different experimental or computational representations of the same protein.
  • Structural archives have limitations that researchers must recognize. Experimental structures may omit flexible regions, contain engineered mutations, or represent molecules outside their native cellular environment. Different structures of the same protein may show different conformations because of ligands, pH, temperature, crystal packing, or other conditions. Predicted structures may be uncertain in flexible or disordered regions, and structural models of complexes may not accurately represent physiological interactions. Consequently, structural information should be interpreted in conjunction with sequence data, functional annotations, experimental evidence, and the biological question being investigated.
  • The Protein Data Bank and related structural archives form a central foundation for modern structural bioinformatics. The PDB provides a major archive of macromolecular structures, wwPDB partners support global access and stewardship, BMRB preserves NMR-related data, EMDB stores electron microscopy maps, and SASBDB provides small-angle scattering data. Predicted structure resources such as AlphaFold Database expand structural coverage beyond experimentally solved molecules. Together, these resources enable researchers to explore molecular architecture, investigate protein function, study interactions, support drug discovery, and connect structural information with the broader landscape of biological data.
Author:

Leave a Reply

Your email address will not be published. Required fields are marked *