Abstract
The Protein Data Bank (PDB) constitutes a collection of the available atomic models of macromolecules and their complexes obtained by various methods used in structural biology, but chiefly by crystallography. It is an indispensable resource for all branches of science that deal with the structures of biologically active molecules, such as structural biology, bioinformatics, the design of novel drugs etc. Since not all users of the PDB are familiar with the methods of crystallography, it is important to present the results of crystallographic analyses in a form that is easy to interpret by nonspecialists. It is advisable during the submission of structures to the PDB to pay attention to the optimal placement of molecules within the crystal unit cell, to the correct representation of oligomeric assemblies and to the proper selection of the space-group symmetry. Examples of significant departures from these principles illustrate the potential for the misinterpretation of such suboptimally presented crystal structures.