Introducing glycomics data into the Semantic Web
- Kiyoko F Aoki-Kinoshita1,
- Jerven Bolleman2,
- Matthew P Campbell3,
- Shin Kawano4,
- Jin-Dong Kim4,
- Thomas Lütteke5,
- Masaaki Matsubara6,
- Shujiro Okuda7, 8,
- Rene Ranzinger9,
- Hiromichi Sawaki10,
- Toshihide Shikanai10,
- Daisuke Shinmachi10,
- Yoshinori Suzuki10,
- Philip Toukach11,
- Issaku Yamada6,
- Nicolle H Packer3 and
- Hisashi Narimatsu10Email author
© Aoki-Kinoshita et al.; licensee BioMed Central Ltd. 2013
Received: 9 May 2013
Accepted: 17 October 2013
Published: 26 November 2013
Glycoscience is a research field focusing on complex carbohydrates (otherwise known as glycans)a, which can, for example, serve as “switches” that toggle between different functions of a glycoprotein or glycolipid. Due to the advancement of glycomics technologies that are used to characterize glycan structures, many glycomics databases are now publicly available and provide useful information for glycoscience research. However, these databases have almost no link to other life science databases.
In order to implement support for the Semantic Web most efficiently for glycomics research, the developers of major glycomics databases agreed on a minimal standard for representing glycan structure and annotation information using RDF (Resource Description Framework). Moreover, all of the participants implemented this standard prototype and generated preliminary RDF versions of their data. To test the utility of the converted data, all of the data sets were uploaded into a Virtuoso triple store, and several SPARQL queries were tested as “proofs-of-concept” to illustrate the utility of the Semantic Web in querying across databases which were originally difficult to implement.
We were able to successfully retrieve information by linking UniCarbKB, GlycomeDB and JCGGDB in a single SPARQL query to obtain our target information. We also tested queries linking UniProt with GlycoEpitope as well as lectin data with GlycomeDB through PDB. As a result, we have been able to link proteomics data with glycomics data through the implementation of Semantic Web technologies, allowing for more flexible queries across these domains.
It is widely acknowledged that developing a mechanism to handle multiple databases in an integrated manner is key to making glycomics accessible to other -omic disciplines. The National Academy of Science published a report called “Transforming Glycoscience: A Roadmap for the Future” that exemplifies the hurdles and problems faced by the Glycomics research community due to the disconnected and incomplete nature of existing databases . Within the last decade, a large number of carbohydrate structure (sequence) databases have become available on the web, all providing their own unique data resources and functionalities . After the conclusion of the CarbBank project , the German Cancer Research Center used the available data to develop their GLYCOSCIENCES.de database , which in general focuses on the three-dimensional conformations of carbohydrates. KEGG GLYCAN was added to the KEGG resources as a new glycan structure database that is linked to their genomic and pathway information . The Consortium for Functional Glycomics also developed a glycan structure database to supplement their data resources storing experimental data from glycan array, glycan profiling from mass spectrometry, glyco-gene knockout mouse and glyco-gene microarray . In Russia, the Bacterial Carbohydrate Structure Database (BCSDB) was developed, which contains carbohydrate structures from bacterial species collected from the scientific literature . Additionally, small databases used in local laboratories have been developed, and so the GlycomeDB database was developed to integrate all the records in these databases to provide a web portal that allows researchers to search across all supported databases for particular structures . The developers of GlycomeDB were a part of the EUROCarbDB project, which was an EU-funded initiative for developing a framework for storing and sharing experimental data of carbohydrates . Several resources were developed under the EUROCarbDB framework including, a database for organizing monosaccharide information was developed, called MonosaccharideDB  and the HPLC-focused database GlycoBase . MonosaccharideDB is an important database for integrating carbohydrate structures from different resources, since oftentimes different representations are used for the same monosaccharides. Unfortunately, funding-support for the EUROCarbDB project ended, however the data resources and software, which are all available as open source software, were taken on by the UniCarbKB project . Meanwhile in Japan, the Japan Consortium for Glycobiology and Glycotechnology Database (JCGGDB) was developed to integrate all the carbohydrate resources in Japan . However, despite all of these efforts to develop useful and valuable glycomics databases, a lack of interoperability is hampering the development of ‘mashup’ applications that are capable of integrating glycan related data with other -omics data.
Almost all databases mentioned above provide their information using web pages restricting the query possibilities to the limited search options provided by the developers. In addition only a few databases provide web services that allow retrieval of data in a machine-readable non-HTML format. The few implemented web service interfaces return proprietary non-standard formats making it hard to retrieve and integrate data from several resources into a single result. Despite some efforts to standardize and exchange their data [14, 15], most glycomics databases are still regarded as “disconnected islands” . Standardization of carbohydrate primary structures is more difficult than genomics or proteomics, mainly because of the inherent structural complexity of oligosaccharides exemplified by complex branching, glycosidic linkages, anomericity and residue modifications. Individual databases developed their own formats to cope with these problems and encode glycan primary structures in a machine readable way .
In order to integrate data in the life sciences using RDF (Resource Description Framework), several annual BioHackathons (Biology + Hacking + Marathon) sponsored by the National Bioscience Database Center (NBDC) and Database Center for Life Science (DBCLS) in Japan have been held since 2008. The 5th BioHackathon was held in Toyama city, Japan, from September 2nd to 7th, 2012 . The glycan RDF subgroup convened in Toyama to discuss and implement the initial version of a contextualized RDF document (GlycoRDF) representing the respective glycan database contents in a standardized RDF format.
For a better understanding of the processes that glycans are involved in, the participants all agreed that not only should the information on primary structures be available but also associated metadata such as the biological contexts the glycans have been found in (including information on the proteins that glycans are linked to), specification of glycan-binding proteins, associated publications and experimental data must be taken into consideration. Such data are spread over the various resources, which are (e.g. in the context of proteins) not limited to only glyco-related databases. A better integration of all these data collections will allow researchers to answer more complex biological questions than simply using individual databases or only cross-linking primary structures. Connecting glycomics resources with other kinds of life science data will also significantly improve the integration of glycan information into systems biology approaches.
Proof-of-concept SPARQL queries
RDFized glycan databases in this study
Number of entries as of May 2013
Number of triples
Over 3300 glycan structures, approximately 9000 structure and protein associations, with over 900 publications
1977 triples for structure and protein data of one experiment
Over 10,000 structures, over 4000 publications, over 5000 taxons, and over 2500 NMR spectra
2,595,411 triples from all data
37,140 glycan structure entries
518,733 triples from all data
About 700 monosaccharide entries
1911 triples of Basetypes, 14,692 triples of Monosaccharides, and 275 triples of Substituents
174 glycoepitopes recognized by 613 antibodies and a wide range of biochemical information related to the glycoepitopes and antibodies
220,545 triples from all data
1,830 entries of mice glycoproteins and 701 of C. elegans glycoproteins
LfDB (Lectin frontier DataBase)
479 entries of lectin data, including PDB information, and their glycan interaction data
902 triples of lectin-PDB relationship data
After the conversion of these data into RDF, we set up a local triplestore using Virtuoso , uploaded all of the data and tested the following queries to see if the target data could be retrieved:
Discussion and conclusion
In this report, we illustrate the utility of RDFizing glyco-databases in order to link glycan data from different glycomics resources with proteomics data. The developers of existing databases agreed upon using RDF as a straightforward approach to link relevant data with one another. This would in turn enable the creation of links with other -omics data sources. In particular, we have shown in this work that the availability of formalized RDF data of glycoscience resources has allowed not only the integrated query of multiple glyco-related databases, but also the integration with UniProt, which is a valuable resource of proteomics data. Although few genomic resources are currently on the Semantic Web, as the utility of this new technology spreads, we expect that other proteomics, metabolomics and even medical data will become available. Moreover, it is a simple matter of adding triples to existing data to link with new resources as they become available, illustrating the power of the Semantic Web.
In order to further add other pertinent glycomics data to the Semantic Web, two points should be kept in mind: 1) the consistent usage of predicates throughout the related data, and 2) the consistent usage of URIs. For 1), it will be necessary to develop an ontology for glycomics data, which is currently under development. For 2), we suggest the usage of identifiers.org when referring to external databases. This base URI is intended to be a persistent URI for any major data resource such that if the original URI changes, identifiers.org will point to the updated resource. Thus users will not need to manage the update of outdated URIs.
Future work entails the development of a more formalized glyco-ontology in order to organize the semantics of the existing glyco-related data, as mentioned above. This can be most easily undertaken by first focusing on the RDF data at hand. As evident from queries 2 and 3, we were forced to use regular expression filters in order to obtain our target data. Thus, we are currently discussing the first version of this glyco-ontology and plan on implementing a more standardized version of our RDF data. This data will be made available as a public SPARQL endpoint in the near future such that federated queries can be performed. This will also make it possible for developers of other related databases to use our standard to most efficiently link their data with the glycomics world.
aNote that in this manuscript, we may use the terms “carbohydrate structure” and “glycan” or “glycan structure” interchangeably. Note also that terms starting with “glyco-“ refer to glycans, which are composed of monosaccharides. For example, glycoproteins are glycosylated proteins, which are protein structures with at least one monosaccharide attached to one of its amino acids.
This work has been supported by National Bioscience Database Center (NBDC) of Japan Science and Technology Agency (JST), National Institute of Advanced Industrial Science and Technology (AIST) in Japan, and the Database Center for Life Science (DBCLS) in Japan. The developers recognize the invaluable contributions from the community and those efforts to curate and share structural and experimental data collections. MC acknowledges funding from the Australian National eResearch Collaboration Tools and Resources project (NeCTAR). PT acknowledges funding from Russian Foundation for Basic Research, grant 12-04-00324. RR is supported by NIH/NIGMS funding the National Center for Glycomics and Glycoproteomics (8P41GM103490).
- Committee on Assessing the Importance and Impact of Glycomics and Glycosciences, Board on Chemical Sciences and Technology, Board on Life Sciences, Division on Earth and Life Studies, National Research Council: Transforming Glycoscience: A Roadmap for the Future. 2012, Washington, D.C., USA: The National Academic PressGoogle Scholar
- Aoki-Kinoshita KF: Using databases and web resources for glycomics research. Mol Cell Proteomics. 2013, 12: 1036-1045. 10.1074/mcp.R112.026252.View ArticleGoogle Scholar
- Doubet S, Albersheim P: CarbBank. Glycobiology. 1992, 2: 505-View ArticleGoogle Scholar
- Lütteke T, Bohne-Lang A, Loss A, Goetz T, Frank M, von der Lieth CW: GLYCOSCIENCES: de: an Internet portal to support glycomics and glycobiology research. Glycobiology. 2006, 16: 71R-81R. 10.1093/glycob/cwj049.View ArticleGoogle Scholar
- Hashimoto K, Goto S, Kawano S, Aoki-Kinoshita KF, Ueda N, Hamajima M, Kawasaki T, Kanehisa M: KEGG as a glycome informatics resource. Glycobiology. 2006, 16: 63R-70R. 10.1093/glycob/cwj010.View ArticleGoogle Scholar
- Raman R, Venkataraman M, Ramakrishnan S, Lang W, Raguram S, Sasisekharan R: Advancing glycomics: implementation strategies at the consortium for functional glycomics. Glycobiology. 2006, 16: 82R-90R. 10.1093/glycob/cwj080.View ArticleGoogle Scholar
- Toukach PV: Bacterial carbohydrate structure database 3: principles and realization. J Chem Inf Model. 2011, 51: 159-170. 10.1021/ci100150d.View ArticleGoogle Scholar
- Ranzinger R, Herget S, von der Lieth CW, Frank M: GlycomeDB-a unified database for carbohydrate structures. Nucleic Acids Res. 2011, 39: D373-D376. 10.1093/nar/gkq1014.View ArticleGoogle Scholar
- von der Lieth CW, Freire AA, Blank D, Campbell MP, Ceroni A, Damerell DR, Dell A, Dwek RA, Ernst B, Fogh R, Frank M, Geyer H, Geyer R, Harrison MJ, Henrick K, Herget S, Hull WE, Ionides J, Joshi HJ, Kamerling JP, Leeflang BR, Lütteke T, Lundborg M, Maass K, Merry A, Ranzinger R, Rosen J, Royle L, Rudd PM, Schloissnig S: EUROCarbDB: An open-access platform for glycoinformatics. Glycobiology. 2011, 21: 493-502. 10.1093/glycob/cwq188.View ArticleGoogle Scholar
- Luetteke T, Monosaccharide DB: http://www.monosaccharidedb.org/ (accessed August 18, 2013)
- Campbell MP, Royle L, Radcliffe CM, Dwek RA, Rudd PM: GlycoBase and autoGU: tools for HPLC-based glycan analysis. Bioinformatics. 2008, 24: 1214-1216. 10.1093/bioinformatics/btn090.View ArticleGoogle Scholar
- Campbell MP, Hayes CA, Struwe WB, Wilkins MR, Aoki-Kinoshita KF, Harvey DJ, Rudd PM, Kolarich D, Lisacek F, Karlsson NG, Packer NH: UniCarbKB: putting the pieces together for glycomics research. Proteomics. 2011, 11: 4117-4121. 10.1002/pmic.201100302.View ArticleGoogle Scholar
- Japan consortium for glycobiology and glycotechnology database.http://jcggdb.jp/index_en.html,
- Packer NH, von der Lieth C-W, Aoki-Kinoshita KF, Lebrilla CB, Paulson JC, Raman R, Rudd P, Sasisekharan R, Taniguchi N, York WS: Frontiers in glycomics: bioinformatics and biomarkers in disease: an NIH white paper prepared from discussions by the focus groups at a workshop on the NIH campus, Bethesda MD (September 11–13, 2006). Proteomics. 2008, 8: 8-20. 10.1002/pmic.200700917.View ArticleGoogle Scholar
- Toukach P, Joshi H, Ranzinger R, Knirel Y, von der Lieth CW: Sharing of worldwide distributed carbohydrate-related digital resources: online connection of the bacterial carbohydrate structure data base and GLYCOSCIENCES.de. Nucleic Acid Res. 2007, 35: D280-D286. 10.1093/nar/gkl883.View ArticleGoogle Scholar
- BioHackathon. 2012, http://2012.biohackathon.org/] (will replace to Biohackathon 2011/2012 paper
- Kaji H, Shikanai T, Sasaki-Sawa A, Wen H, Fujita M, Suzuki Y, Sugahara D, Sawaki H, Yamauchi Y, Shinkawa T, Taoka M, Takahashi N, Isobe T, Narimatsu H: Large-scale identification of N-glycosylated proteins of mouse tissues and construction of a glycoprotein database, GlycoProtDB. J Proteome Res. 2012, 11: 4553-4566. 10.1021/pr300346c.View ArticleGoogle Scholar
- Lectin Frontier DataBase.http://jcggdb.jp/rcmg/glycodb/LectinSearch,
- Orri E, Mikhailov I: RDF Support in the Virtuoso DBMS. Conference on Social Semantic Web. 2007, 113: 59-68.Google Scholar
- Consortium UP: Update on activities at the universal protein resource (UniProt) in 2013. Nucleic Acids Res. 2013, 41: D43-D47.View ArticleGoogle Scholar
- Herget S, Ranzinger R, Maass K, Lieth CW: GlycoCT-a unifying sequence format for carbohydrates. Carbohydr Res. 2008, 343: 2162-2171. 10.1016/j.carres.2008.03.011.View ArticleGoogle Scholar
- Berman HM, Westbrook J, Feng Z, Gilliland G, Bhat TN, Weissig H, Shindyalov IN, Bourne PE: The protein data bank. Nucleic Acids Res. 2000, 28: 235-242. 10.1093/nar/28.1.235.View ArticleGoogle Scholar
- Lutteke T, Frank M, von der Lieth CW: Data mining the protein data bank: automatic detection and assignment of carbohydrate structures. Carbohydr Res. 2004, 339: 1015-1020. 10.1016/j.carres.2003.09.038.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.