Automatic construction of gene relation networks using text mining and gene expression data

Author： Karopka Thomas Scheel Thomas Bansemer Sven Glass Änne

ISSN： 1464-5238

Source： Medical Informatics and the Internet in Medicine, Vol.29, Iss.2, 2004-06, pp. : 169-183

Disclaimer: Any content in publications that violate the sovereignty, the constitution or regulations of the PRC is not accepted or approved by CNPIEC.

Previous Menu Next

Abstract

Microarray gene expression analysis is a powerful high-throughput technique that enables researchers to monitor the expression of thousands of genes simultaneously. Using this methodology huge amounts of data are produced which have to be analysed. Clustering algorithms are used to group genes together based on a predefined distance measure. However, clustering algorithms do not necessarily group the genes in a biological meaningful way. Additional information is needed to improve the identification of disease relevant genes. The primary objective of our project is to support the analysis of microarray gene expression data by construction of gene relation networks (GRNs). Required information can not be found in a structured representation like a database. In contrast, a large number of relations are described in biomedical literature. The main outcome of this project is the implementation of a software system that provides clinicians and researchers with a tool that supports the analysis of microarray gene expression data by mapping known relationships from the biomedical literature to local gene expression experiments.