BMRB/XML & BMRB/RDF Data Server:

common open representations of BMRB NMR-STAR data in XML and RDF formats

This site provides BMRB NMR-STAR data derived from NMR spectroscopy in XML and RDF formats. BMRB/XML and BMRB/RDF are names of data collections, respectively.

Data are updated every Thursday. Last update: Oct 28, 2016

BioMagResBank (BMRB) is a repository for experimental and derived data gathered from nuclear magnetic resonance (NMR) spectroscopic studies of biological molecules. As a member of the Worldwide Protein Data Bank (wwPDB), the BMRB maintains archives of quantitative NMR spectral parameters (e.g. assigned chemical shifts, J-coupling constants, etc.), derived data (e.g. relaxation parameters, kinetics parameters, NMR restraints used to determining the structures), time-domain spectral data (FID) and a metabolite NMR spectral database. The BMRB provides data deposition, validation and visualization tools in collaboration with other wwPDB members (RCSB-PDB in USA, PDBj in Japan, PDBe in Europe).

Representing an alliance of BMRBs (USA at University of Wisconsin-Madison, Japan at Osaka University and Europe at University of Florence), PDBj-BMRB provides the NMR-STAR data in web standard formats, XML and RDF to enhance the interoperability of the BMRB archival data.

Schema of the BMRB/XML (BMRB/XML Schema) is translated from the current NMR-STAR v3.1 Dictionary and make it easy for scientists familiar with NMR-STAR files to comprehend the contents in the XML format. BMRB/XML accommodates a rich information content in a single file including entry citations, experimental parameters, assigned chemical shifts, links to external databases, atom coordinates of the biomolecules, NMR restraints, assigned peak lists, etc.

Both BMRB/XML and BMRB/RDF are based on NMR-STAR v3.1.1.65 Dictionary.
We provide two versions of the BMRB/XML; "complete" and "noatom" versions, respectively. The "noatom" version does not include the atomic coordinate, NMR restraints and peak lists for ease of handling.

The BMRB/RDF files are generated by direct translation from the corresponding BMRB/XML data files. Along with the BMRB/RDF, we provide its ontology of the BMRB/RDF (BMRB/OWL) in Web Ontology Language (OWL). The RDF/RDFS/OWL Semantic Web flamework constitute a standard model for data exchange on the Web, and make it possible to query across diverse data sources using SPARQL.

We provide only "noatom" version of the BMRB/RDF.

NMR-STAR format family

As all three formats, NMR-STAR, BMRB/XML and BMRB/RDF, are consistently based on the current NMR-STAR v3.1 Dictionary forming a family of NMR-STAR data formats, the information content represented by each is preserved.

Ontology Canonical NMR-STAR v3.1 Dictionary
Extended dictionary for BMRB/XML
Database DDL
DDL based on BMRB/XML Schema
DDL compatible with NMR-STAR Dic.
Validation ADIT-NMR, Manual annotation using in-house validator in-house DB loader Automated remediation by BMRBxTool
Full XML Schema validation by Apache Xerces
RDF validation by BMRBoTool with Redland Raptor
Data type check ADIT-NMR, Manual annotation using in-house validator
(string based)
Database schema
(DDL based)
XML Schema data types
(ID, string, integer, real, date, enumeration)
Metadata check referring to external DBs
(PubMed, Taxonomy, PDB, Ligand Expo, etc.)
Inheriting all XML Schema data types
Type of storage Text files (.str) RDB XML files (.xml), XML Database RDF files (.rdf, .nt), RDF triplestore
Standardization BMRB BMRB BMRB W3C BMRB W3C
Parser libraries
starlibj, starlib2 and SANS
JDBC, ODBC drivers and various language bindings DOM, SAX, StAX and JAXB libraries and various language bindings RDF parsing libralies, RDF triplestore
Query languages Not standardized, but you can do with either starlibj, starlib2 or SANS SQL XPath, XSLT, XQuery SPARQL
Information content Canonical repositories:
Extended repositories:
LACS validation report
Protein Blocks annotation
CS Completeness
Conventional + "BMRB+PDB" + PACSY
and Metabolomics
link to PACSY structural annotation
"complete" and "noatom"
"complete" accommodates the next repositories:
Conventional, BMRB+PDB, PACSY, LACS, Protein Blocks, CS completeness.
"noatom" omits coordinate, restraints and peak lists from the "complete".
RDF/XML or N-Triples
Both accomodate information content of "noatom" version of BMRB/XML.

Comparison of information contents

The next Euler diagram showing that the extended NMR-STAR Dictionary consists of the canonical NMR-STAR Dictionary and extra definitions about related data repositories, LACS chemical shift validation report, Protein Blocks structural annotation and Completeness of assigned chemical shifts. It also shows relationship between the extended NMR-STAR Dictionary and the BMRB/XML Schema, having one-to-one correspondence. Thus, the BMRB/XML ("complete") would be the most comprehensive NMR-STAR data repository as a single format.
Information content

Linked data with BMRB

The next graph represents linked external information resources via RDF links, where shorter distances from BMRB indicate closer relationships with BMRB.
Linked Data


Yokochi M, Kobayashi N, Ulrich EL, Kinjo AR, Iwata T, Ioannidis YE, Linvy M, Markley JL, Nakamura H, Kojima C, Fujiwara T, Publication of nuclear magnetic resonance experimental data with semantic web technology and the application thereof to biomedical research of proteins, J. Biomed. Semantics, 7, 1-4 (2016)

Ulrich EL, Akutsu H, Doreleijers JF, Hrano Y, Ioannidis YE, Lin J, Linvy M, Mading S, Maziuk D, Miller Z, Nakatani E, Schulte CF, Tolmie DE, Wenger RK, Yao H, Markley JL, BioMagResBank, Nucleic Acids Research, 36, D402-D408 (2008)