=Paper=
{{Paper
|id=Vol-1114/Poster_Katayama
|storemode=property
|title=Interactive Platform for Semantic Gene Expression Analysis of Alzheimer's Disease
|pdfUrl=https://ceur-ws.org/Vol-1114/Poster_Katayama.pdf
|volume=Vol-1114
|dblpUrl=https://dblp.org/rec/conf/swat4ls/KatayamaKO13
}}
==Interactive Platform for Semantic Gene Expression Analysis of Alzheimer's Disease==
https://ceur-ws.org/Vol-1114/Poster_Katayama.pdf
Interactive platform for semantic gene expression
analysis of Alzheimer’s disease
Toshiaki Katayama1, Masataka Kikuchi2, Soichi Ogishima3
1
Database Center for Life Science, Research Organization of Information and Systems,
Tokyo, Japan
ktym@dbcls.jp
2
Department of Molecular Genetics, Center for Bioresources, Brain Research Institute,
Niigata University, Niigata, Japan
kikuchi@bri.niigata-u.ac.jp
3
Department of Bioclinical Informatics, Tohoku Medical Megabank Organization, Tohoku
University, Miyagi, Japan
ogishima@megabank.tohoku.ac.jp
Abstract. In the course of gene expression analysis, it is required to interpret
data by referencing knowledge bases of genetics, pathways, diseases and drugs.
However, because those external resources are often stored in distributed
databases in various formats, it is hard for biomedical scientists to use them in
combination. Semantic Web technologies are suitable for integration of those
heterogeneous datasets using Resource Description Framework (RDF) and
providing a faceted search interface. In this work, we report an application of
semantic data integration with faceted search for interactive analysis of gene
expression data of Alzheimer’s disease. This framework can be easily extended
for other biomedical domains.
Keywords: data integration, linked open data, faceted search, gene expression
analysis, Alzheimer’s disease
Introduction
Difference of gene expressions of target and control samples can be measured by
microarray technology. Those data are primarily processed to find genes having
excessive fold change with a significant probability. Interpretation of relationship
between those candidate genes and a hypothesis behind a design of the experiment
depends on supportive information from existing knowledge. This includes genetics,
pathways, diseases and drugs which are independently stored in different databases
and only available in incompatible data formats. For integration of heterogeneous
datasets, it is becoming standard to use RDF in life sciences [1]. With semantic Web
technologies, relations of data can be represented as a linked graph which provides
powerful means to explore a set of related information often referred to as faceted
2 Toshiaki Katayama1, Masataka Kikuchi2, Soichi Ogishima3
search. Here we report an application which can be used for exploring gene
expression data with integrated semantic datasets for Alzheimer’s disease.
Results
We introduced RDF for data integration of Affymetrix microarray probes, Human
Genome Organization (HUGO) gene annotations, UniProt protein annotations, gene-
drug interactions in DrugBank, disease susceptibility genes in AlzGene [2] and
molecular interactions in AlzPathway [3] databases for Alzheimer’s disease. Most of
those datasets except for UniProt are converted into RDF by in-house developed
scripts. Based on these datasets stored in a triple store, we developed a new Web
application (Fig. 1) that accepts gene expression data and provides interactive
interface to explore the list of candidate genes with various facets including gene
expression ratio, chromosomal positions of genes, protein interactions with drugs,
known genetic diseases, and pathway information.
Figure 1. Workflow of interactive faceted analysis.
The server is implemented in Node.js which accesses a backend triple store to
perform faceted search by SPARQL Protocol and RDF Query Language (SPARQL).
The Web interface is developed using D3.js, Bootstrap and Google Map API.
Interactive platform for semantic gene expression analysis of Alzheimer’s disease 3
Discussions
Our application is designed for exploring candidate genes of Alzheimer’s disease
based on the user-supplied gene expression data by referencing existing knowledge
with a faceted navigation. We found that Semantic Web technologies are well-suited
for data integration and implementation of a faceted search interface. This means that
the method and the interface we developed can be extended for any other biomedical
domains which requires integration of heterogeneous data where interactive search
can assist a data-driven approach.
References
1. Katayama T, Wilkinson MD, Micklem G, et al.: The 3rd DBCLS BioHackathon:
improving life science data integration with semantic Web technologies. Journal
of Biomedical Semantics 2013, 4:6.
2. Bertram L, McQueen MB, Mullin K, Blacker D, Tanzi RE: Systematic meta-
analyses of Alzheimer disease genetic association studies: the AlzGene database.
Nature genetics 2007, 39:17–23.
3. Mizuno S, Iijima R, Ogishima S, et al.: AlzPathway: a comprehensive map of
signaling pathways of Alzheimer’s disease. BMC systems biology 2012, 6:52.