=Paper=
{{Paper
|id=Vol-1546/poster_24
|storemode=property
|title=RDF-Based Integration with SPARQL Building System for Life Science Database Archive
|pdfUrl=https://ceur-ws.org/Vol-1546/poster_24.pdf
|volume=Vol-1546
|authors=Atsuko Yamaguchi,Katsuhiko Okubo,Norio Kobayashi,Kouji Kozaki,Sadahiro Kumagai,Kai Lenz,Tomoe Nobusada,Hongyan Wu,Yasunori Yamamoto,Hideki Hatanaka
|dblpUrl=https://dblp.org/rec/conf/swat4ls/YamaguchiOKKKLN15
}}
==RDF-Based Integration with SPARQL Building System for Life Science Database Archive==
https://ceur-ws.org/Vol-1546/poster_24.pdf
RDF-Based Integration with SPARQL Building
System for Life Science Database Archive
Atsuko Yamaguchi1 , Katsuhiko Okubo2 , Norio Kobayashi3 , Kouji Kozaki4 ,
Sadahiro Kumagai2 , Kai Lenz3 , Tomoe Nobusada5 , Hongyan Wu1 , Yasunori
Yamamoto1 , and Hideki Hatanaka5
1
Database Center for Life Science (DBCLS), ROIS,
178-4-4 Wakashiba, Kashiwa, Chiba, 277-0871 Japan
{atsuko, wu, yy}@dbcls.rois.ac.jp
2
Information & Telecommunication Systems Company, Hitachi Ltd,
6-26-2 Minami Oi, Shinagawa-ku, Tokyo 140-8573 Japan
{katsuhiko.okubo.yh, sadahiro.kumagai.jj}@hitachi.com
3
Advanced Center for Computing and Communication (ACCC), RIKEN,
2-1 Hirosawa, Wako, Saitama, 351-0198 Japan
{norio.kobayashi, kai.lenz}@riken.jp
4
The Institute of Scientific and Industrial Research (ISIR), Osaka University,
8-1 Mihogaoka, Ibaraki, Osaka, 567-0047 Japan
kozaki@ei.sanken.osaka-u.ac.jp
5
National Bioscience Database Center, JST,
5-3, Science Plaza 7F, Yonbancho, Chiyoda-ku, Tokyo 102-8666 Japan
{nobusada, hideki}@biosciencedbc.jp
Abstract. The Life Science Database Archive (LSDB Archive,
https://dbarchive.biosciencedbc.jp/ ) is a service to collect, preserve and
provide databases generated by life-science researchers in Japan. As of
September 2015, the LSDB Archive includes 103 databases and all the
databases can be downloadable with appropriate licenses and metadata.
Although a simple keyword search tool is available for the databases,
more flexible retrieval system to obtain relevant data from heteroge-
neous databases is required. Therefore, we first converted the databases
into RDF datasets, uploaded in a triple store. Then, we developed a
prototype of retrieval system using SPARQL Builder. Because SPARQL
Builder assists users in writing queries, the prototype enables users with-
out knowledge of RDF to access the datasets.
Keywords: SPARQL, RDF, database archive, life-science databases
1 Introduction
In life sciences, many kinds of data have been generated as results of experi-
mental research. Although they are often organized and provided as databases,
we found that many databases in Japan were not maintained anymore after
fundings of projects end as of 2006 [1]. Even in case that databases are main-
tained, because they were unclear with respect to the terms of use, or were not
downloadable, they may not be fully used. To address these issues, the Life Sci-
ence Database Archive (LSDB Archive, https://dbarchive.biosciencedbc.jp), a
service to maintain, store and provide downloadable databases with appropriate
licences and unified descriptions, started from 2009. As of September 2015, the
LSDB Archive includes 103 heterogeneous databases generated by life-science
researchers during national projects in Japan.
Although a simple keyword search tool is available for the databases, more
flexible retrieval system to obtain relevant data from heterogeneous databases
is required. To do so, we decided to use semantic web technologies to integrate
the databases. We converted the databases into RDF datasets, and uploaded in
a triple store with a SPARQL endpoint as a trial. However, a SPARQL query
construct is intractable to users who are unfamiliar with semantic web technolo-
gies although a SPARQL endpoint is very flexible retrieval system. As a system
for assisting users to write SPARQL query, we employed SPARQL Builder [2]
that is a semiautomatic SPARQL query generation system. Using this system,
we developed a prototype of a search interface for LSDB Archive that enable
users to extract semantically related data.
2 Method and Result
To generate initial RDF datasets from databases in LSDB Archive, we used To-
goDB [3], which accepts tabular formatted data and generates RDF datasets.
By indicating a class for each column in TogoDB, rdf:type for objects are au-
tomatically attached. To type subjects, we added an additional data including
rdfs:domain for each property corresponding to a column. Because all the classes
used in the RDF datasets are introduced from external ontologies, we extracted
necessary part of those ontologies. Then we generated SPARQL Builder meta-
data for the RDF datasets together with extracted ontologies.
We then developed a prototype of a search interface on RDFized LSDB
Archive using SPARQL Builder. Using this interface, by selecting two classes and
a relationship between the two classes, users can search for desired data using
SPARQL query from combined heterogenious RDF datasets in LSDB archive.
Acknowledgments. This work was supported by the National Bioscience Data-
base Center (NBDC) of the Japan Science and Technology Agency (JST).
References
1. Ministry of Education, Culture, Sports, Science, and Technology (MEXT) Inte-
grated Database Project: http://lifesciencedb.mext.go.jp/en/
2. Yamaguchi A., Kozaki K., Lenz K., Wu H, Kobayashi N.: An Intelligent SPARQL
Query Builder for Exploration of Various Life-science Databases, CEUR Workshop
Proceedings 1279, The 3rd International Workshop on Intelligent Exploration of
Semantic Data (IESD 2014), Riva del Garda, Italy.
3. TogoDB: http://togodb.org/