STAR Software and Dictionaries For Macromolecular NMR. David Argentar, Anthony D'Silva, Sridevi Bhamidipati, Eldon L. Ulrich, Yannis E. Ioannidis, Miron Livny, and John L. Markley, Departments of Biochemistry and Computer Sciences, University of Wisconsin, Madison, WI 53706

BMRB has developed a series of software tools for creating, reading, and manipulating STAR files [1]. We have also written data dictionaries to record macromolecular NMR data in STAR files (NMR-STAR) and we are maintaining the archive as a collection of these files.

At present, BMRB has developed a C library to allow reading of STAR files. This library is used by BMRB software to extract data from our STAR files and to allow WWW users to search and visualize BMRB data.

Since NMR-STAR is used exclusively to deposit data to BMRB, we have developed software to automate common tasks in depositing data, including adapting the Protein Data Bank's AutoDep system to read and write NMR-STAR. As part of this effort, we have developed tools for translating between STAR and CIF, including methods for emulating STAR features not found in CIF.

Ongoing development work involves writing code to check referential integrity of STAR files and to validate data in STAR files against the complex requirements defined by the NMR-STAR dictionary. A dictionary and software are under development to allow NMR-STAR to be used with the BMRB relational database.

Supported by grant LM05799-01A2 from the National Library of Medicine.

[1] Hall, S. R. and Spadaccini, N., The Star File: Detailed Specifications, J.Chem. Inf. Comput. Sci., Vol. 34, pp. 505-508 (1994).