Microarray Data Analysis: From Preparation To Classification

Microarray is a well-established technology to analyze the expression of many genes in a single reaction whose applications range from cancer diagnosis to drug response. In cancer classification and prediction, microarrays help by analyzing the expression of genes in tumor cells taken from different tissues. This chapter sketches the experimental design and the normalization of microarrays. Next, it surveys the most used statistical tests for the gene ranking. The chapter then focuses on the commonly used classification methods for profiling data, and reviews state-of-the art approaches. It presents a new classification method, called Microarray Interval Discriminant CLASSifier (MIDClass). MIDClass is based on association rules built on top of gene expression intervals. A comparative analysis on a case study of statistical tests is presented. This is followed by wide experimental evaluation on the classification techniques showing the effectiveness of MIDClass compared to the most prominent classification approaches.