Commonly used statistical methods for detecting differential gene expression in microarray experiments

Bioinformatics tools provide the needs for carrying out microarray analysis. Differential gene expression analysis reveals the substantial aspects of biological pathways and takes a leading part for further hypothesis development. The Principle in analyzing the gene expression data is the need for determination the genes of which expression models differ by phenotype or an experimental condition. Microarrays are employed to detect the distinctive gene expression profiles in a wide variety of experimental conditions. In the field of bioinformatics, different computer science technologies and statistical methods are considered. This multidisciplinary approach allows the understanding of the relationship between statistical methods, bioinformatics applications and computer science technologies. Images resulting from microarray experiments are not directly interpreted to reveal differences in sample condition. To make microarray experiments interpretable, a number of algorithms and statistical approaches need to be applied. The raw dataset has to go through preprocessing step prior to a series of subsequent analyses as significance analysis clustering and visualization of the relevant biological components and samples. Many microarray studies are designed to detect genes associated with different phenotypes. This review attempts to give an overview of statistical methods for identifying differentially expressed genes in microarray experiments. In this review, our aim is to summarize briefly basic methods as a guide for differential gene expression studies. All the summarized methods are given from the basic to more complicated in the review.