Protein Data Bank

The Protein Data Bank (PDB) is a repository for 3-D structural data of proteins and nucleic acids. This data, typically obtained by X-ray crystallography or NMR spectroscopy, is submitted by biologists and biochemists from around the world, is released into the public domain, and can be accessed for free. The database is the central repository for biological structural data.

History

Founded in 1971 by Brookhaven National Laboratory, the Protein Data Bank was transferred in 1998 to the Research Collaboratory for Structural Bioinformatics (RCSB), which is composed of Rutgers University, the University of Wisconsin, Madison, NIST and the San Diego Supercomputer Center. Funding comes from the National Science Foundation, Department of Energy, National Library of Medicine and the National Institute of General Medical Sciences. The European Bioinformatics Institute in the UK and the Institute for Protein Research in Japan also collect, process and submit data files.

The PDB is a key resource in structural biology and is critical to more recent work in structural genomics.

Countless derived databases and projects have been developed to integrate and classify the PDB interms of protein structure, protein function and protein evolution.

As of 1 October, 2004, the database contained 27,428 released atomic coordinate entries (or "structures") and took in about 2,000-3,000 new ones per year. Data are stored in the mmCIF format specifically developed for the purpose.
Note that the database stores information about the exact location of all atoms in a large biomolecule; if one is only interested in sequence data, i.e. the list of amino acids making up a particular protein or the list of nucleotides making up a particular nucleic acid, the much larger databases from Swiss-Prot and the International Nucleotide Sequence Database Collaboration should be used.

Statistics

As of 22 February, 2005, the "PDB Holdings List" at RCSB reported the following statistics:

Through the years the PDB has undergone many, many changes and revisions. Its original format was dictated by the width of computer punch cards.

RCSB PDB Guide The PDB format specification can be found here, and it is vital that you read this before looking at the raw data.
ftp.rcsb.org The raw data can be downloaded from here.
www.rcsb.org Statistics about the PDB can be found here.

This legacy format has caused many problems with the format, and consequently the PDB has three distinct 'clean-up' projects;

The Molecular Modeling DataBase (MMDB) from NCBI

The Macromolecular Structure Database from the European Bioinformatics Institute

The Data Uniformity Project from PDB

Each of these grant-funded projects has attempted to achieve the same goal via different routes. The Data Uniformity Project is hosted by the RCSB (the current home of the PDB). Each uses the original PDB data to derive a new format; The MMDB uses ASN.1 (and an XML conversion of this format); The MSD uses a Relational Database; The Data Uniformity Project uses mmCIF (and another XML conversion of this format).

Some people would say that this is a Good Thing; others would argue that, without a universal repository of information (i.e., a common dictionary), how can we talk about the same thing.

Each structure published in PDB receives a four-character alphanumeric identifier, its PDB ID. This should not be used as an identifier for biomolecules, since often several structures for the same molecule (in different environments or conformations) are contained in PDB with different PDB IDs.

If a biologist submits structure data for a protein or nucleic acid, PDB staff reviews and annotates it. The data are then automatically checked for plausibility. The source code for this validation software has been released for free. The main data base accepts only experimentally derived structures, and not theoretically predicted ones (see protein structure prediction).

Various funding agencies and scientific journals now require scientists to submit their structure data to PDB.

Viewing the data

The structural data can be used to visualize the biomolecules with appropriate software, such as rasmol, chime or a web browser VRML plugin.
The PDB website also contains resources for education, structural genomics, and related software.

Links to enzyme database data

[2] The best mapping is provided by Kim Henrick's group at EBI.

[1] PDB provide a mapping on their beta site, but it is at the whole PDB level not chain level.

[1] Search at BRENDA enzyme database portal.

[1] PDBSProtEC:

References

Bernstein FC, Koetzle TF, Williams GJ, Meyer Jr EF, Brice MD, Rodgers JR, Kennard O, Shimanouchi T, Tasumi M. The Protein Data Bank: a computer-based archival file for macromolecular structures. J Mol Biol 1977;112:535-542. PMID 875032.