User:Zuraini A. Shah

Feature selection and classification on gene expression data
ABSTRACT Gene expression data has been discovered to support many important disciplines such as medical, pharmacies and biological. Nowadays, the analysis from gene expression data has encouraged biologist to studies a critical disease that attacks a human such as cancer. In clinical application, a few studies tend to use molecular gene expression profiling to overcome the limitation of clinical predictors that are not accurate enough for prediction in the individual patient. However, using whole of the features for prediction task is impractical and can lead to reduce the diagnosis accuracy. One of the factors that lead to this problem is because the gene expression data contain huge irrelevant information. Meanwhile, by selected the significant or informative features can help the biologists and doctors make a further study especially for finding the relation between disease-gene.

INTRODUCTION DNA microarray technology has provided biologists with the capability to measure the level of thousands of genes in a single experiment. One of the major current applications of microarray technology is using the genome-wide expression data in order to classify samples taken from different tissues. This application can be deployed for diagnosis purpose such as classifying cancer tissues in order to predict the status or the type of the disease. Thus, using gene expression profile in clinical application brings a few advantage such as studying the gene-disease relation and diagnosis application. Based on the background problem, this research is categorized as a feature selection and classification problem where gene expression analysis as a domain. Due to this assumption, there are three main processes that are dominant in this research. They are feature selection, classification and biological validation.

Education

 * 1) Expected 2010, PhD in Computer Science (Research in Bioinformatics), Universiti Teknologi Malaysia.
 * 2) 1999, MSc in Computer Science, Universiti Teknologi Malaysia.
 * 3) 1997, BSc in Computer Science (Software Engineering), Universiti Teknologi Malaysia.

Research Interests

 * 1) Gene expression analysis
 * 2) Feature Selection
 * 3) Classification
 * 4) Clustering

Group Members

 * 1) Dr Razib M. Othman, Laboratory of Computational Intelligence and Biology (LCIB)
 * 2) Pn Shahreen Kasim, Laboratory of Computational Intelligence and Biology (LCIB)

Contact Info
Zuraini Ali Shah Laboratory of Computational Intelligence and Biology (LCIB) No. 204, Level 2 Industry Centre, Technovation Park Universiti Teknologi Malaysia (UTM) 81310 UTM Skudai, MALAYSIA Mobile: +6019-799-8943 Tel/Fax: +607-559-9230 Email: aszuraini@utm.my