Title:	HUMAN-MICROBIOME-RELATIONS EXTRACTION AND VISUALIZATION SYSTEM WITH CONTEXT-DEPENDENT CLUSTERING AND SEMANTIC ANALYSIS
Author(s):	Shiori Hikichi, Shiori Sasaki, Yasushi Kiyoki
ISBN:	978-989-8533-45-6
Editors:	Hans Weghorn
Year:	2015
Edition:	Single
Keywords:	Bacteria, data mining, human gut microbiome, normalization, personalized medicine
Type:	Full Paper
First Page:	105
Last Page:	112
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	Human gut microbiome is a set of bacteria, which is significantly larger number of somatic cells, providing various pathological and biological impacts on a hosting human body system. Recent Next-Generation Sequencing (NGS) enables to perform quantitative analysis on a large scale of human gut microbiome data. Though several studies have investigated the relationship between microbiome behavior and human’s attributes (clinical-pathological parameters), such as national origin, age and diet style, these data are not utilized, analyzed and shared fully because the standardization of system is not achieved sufficiently. Thus, the objective of this research is to realize a new analytical system for human gut microbiome data to extract unknown relationships between human attributes and microbial composition with a context-dependent clustering and semantic analysis methods, and contribute to the effective data utilization and system standardization. The most important feature of our system is to analyze the unknown relations of human-microbiome from individual data in a context-dependent way. With this system, an analyst is able to grasp the overview of bacteria data clustered by human attributes in a scatter-diagram or a dendrogram by selecting human attributes as a set of context, and extract the unknown relations of human-microbiome by data mining methods: people in a country have different bacterial components from in other countries; the high percentage of Bacteroides and the low percentage of Prevotella. This integrated database system with analytical visualization functions will contribute to the advanced personalized medicine from the viewpoint of bacteriology.
	Go Back