The Nucleic Acid Database: Present and Future

Abstract

The Nucleic Acid Database is a relational database containing information about three-dimensional nucleic acid structures. The methods used for data processing, structure validation, database management and information retrieval, as well as the various services available via the World Wide Web, are described. Plans for the future include greater reliance on the Macromolecular Crystallographic Information File for both data processing and data management.

Keywords: Macro-molecular Crystallographic Information File; database management; nucleic acid structure.

Fig 1a

A schematic diagram of NDB data processing that illustrates the central role of the NDB filter software in automating the exchange of information between a variety of input formats and the mmCIF template and data file archival format.

Fig 1b

Schematic view of the data flow in and out of the NDB database as of October 1995. The figure illustrates the generation of derived structural features by the NDBquery program using both internal functions and encapsulated external programs. The collection of report types created by NDBquery is also shown. All of these reports are accessible via the NDB network server.

Fig. 2

Sequence for a simple query, i.e., choosing structures that contain the specific sequence ACGCG using the WWW Interface, version 2.0 (October 1995).

Fig. 3

Examples of Postscript reports created by NDBquery.

Fig. 4

Examples of Postscript graphs created by NDBquery. (a) Histogram of the distribution of the C5′–C4′ bond lengths in Z-DNA. (b) Pie chart showing the distribution of structure types in the NDB. (c) Scatterchart of ζ vs β-torsion angles in successive residues of high resolution B-DNA structures, (d) Conformation wheel of the observed torsion angles in the Dickerson dodecamer, BDL001 [11].

Fig. 5

Examples of Postscript molecular graphics created by NDBquery for the self-complimentary duplex d(CGATCGATCG)₂, BDJ025 [12]. (a) Ball and stick, (b) Stercotriptych [13J. (c) Four representative views, (d) Packing diagrams.

Fig. 6

Torsion angle wheels for the B-DNA structure BDL001 [11]. The expected ranges for A-DNA, B-DNA and Z-DNA are shaded. In this example, all of the values for the torsion angles fall completely within the B-DNA range.

Fig. 7

The NDB Homepage (available at http://ndbserver.rutgers.edu and is mirrored at the European Bioinformatics Institute at http://www.ebt.ac.uk/NDB/).

Fig. 8

The NDB Archives Page.

Fig. 9

An Atlas entry for the first B-DNA crystal structure, BDL001 [13]. (a) The top of the entry page shows the structure type, compound name, sequence, citation, and space group, (b) Also included in the atlas entry are cell constants, crystallization conditions, refinement, and a link to the coordinate file for the structure. A ball and stick representation of the structure is color coded by sequence, with thymine in blue, adenine in red, cytosine in yellow, and guanine in green. (c) The space filling and ribbon representations of the unit cell are color coded in terms of the symmetry related molecules.

Fig. 10

Functional diagram of SIFLIB illustrating the interaction of this utility library with a variety of other applications. The figure highlights the role of SIFLIB in encapsulating access to CIF Format data and dictionaries from calling applications.

Fig. 11

Schematic view of the NDB WWW forms based interface. The WWW version of NDBquery is called by the WWW server and provides the server with a description of the contents of the NDB database, which is presented as a set of menu selections. The WWW interface also manages the construction of SQL queries and all communication with the NDB database.