Development of PDBj-ML

The methodological advance of protein structure determination has resulted in accumulation of a large amount of structural information of biological macromolecules. With the advent of the structural genomics, the amount is expected to increase even more rapidly in the near future. Efficient methods of its storage and handling are, therefore, required to fully explore the information available. The Protein Data Bank (PDB) [1, 2] has been the primary database of the field, where atomic coordinates and other relevant information are stored in a well-known PDB format as well as the Macromolecular Crystallographic Information Format (mmCIF) [3]. Since last year Institute for Protein Science (IPR) of Osaka University has collected, curated and edited some of those coordinates as PDBjapan (PDBj) in collaboration with Research Collaboratory for Structural Bioinformatics (RCSB). In recent years, XML has been proven powerful in various fields for the exchange and archiving of data, including biology-oriented datebases. It would be useful, therefore, to have an XML version of PDB so that it can be easily combined with other biological databases. In an attempt to provide such an interface between structural data and other tools of bioinformatics, we have initiated development of PDBj-ML, an XML format for structural information of macromolecules.