Tony Smith

Microarray analysis

There is considerable activity and interest within the biological research community towards using microarray analysis as a tool for the prediction of genetically hetereogenious diseases: specifically for predicting prognosis and treatment outcomes for such diseases as cancer (e.g. IOEC). Statistical analysis and dendrograms have proven to deliver some degree of success, but the utility of supervised machine learning algorithms in general has not been explored very much.

This project will explore application of the machine learning algorithms in WEKA to microarray data. Specifically, we will start with the well-documented problem of predicting a prognosis of ovarian cancer from the MC160 data (MC160)

This project is appropriate for students with an interest in the application of machine learning to biological problems. Good JAVA programming skills are essential.

Department of Computer Science, University of Waikato, Hamilton, New Zealand