A Redesigned Database of Three-dimensional Protein Structures for Reuse in Macromolecular Crystallography

The number of macromolecular structures solved and deposited to the Protein Data Bank (PDB) is
more than 66000. Exploiting this information for macromolecular crystallography (MX) dramatically
increases the efficiency of MX structure solution. This paper describes a database of "unique" PDB
entries with their domains and potential oligemeric states. The database is used, as a proof of prin-
ciple, in one of the automatic MX structure solution pipelines. Tests with the new entries from the PDB
shows that the pipeline based on this database is able to solve around 80% of all solvable structures
automatically.