Knowledge discovery in Hepatitis C Virus transgenic mice

Abstract

For the purpose of gene identification, we propose an approach to gene expression data mining that uses a combination of unsupervised and supervised learning techniques to search for useful patterns in the data. The approach involves validation and elimination of irrelevant data, extensive data pre-processing, data visualization, exploratory clustering, pattern recognition and model summarization. We have evaluated our method using data from microarray experiments in a Hepatitis C Virus transgenic mouse model. We demonstrate that from a total of 15311 genes (attributes) we can generate simple models and identify a small number of genes that can be used for future classifications. The approach has potential for future disease classification, diagnostic and virology applications.