ABSTRACT
Producing a useful list of differentially expressed genes from microarray data sounds like an easy task. However, our experience is different. Being caught in various traps ourselves, we will warn of the most dangerous pitfalls. Some ‘roads to the list’ are relatively safe to travel on, and we will point them out. The trip usually splits into two parts, and can be stopped after the first. At the beginning is explorative analysis which leads to candidate genes. Having arrived at this destination, you can transit to statistical analysis. This leads to quality measures, and allows you to distinguish between a reliable result and an unreliable one.
