ABSTRACT
This chapter aims to introduce the clinical investigator to the day-to-day problems encountered in the design, implementation, and analysis of microarray analysis of clinical samples and provides the background necessary to develop solutions to these problems. It discusses the labelling technology for the cDNA and oligo microarrays. The chapter discusses the potential sources of variation, options for decreasing variability, and the applicable statistical tests with relation to generating the most efficient experimental design using clinical samples. There are four main sources of experimental variability: measurement error, technical variation, biological variation, and treatment dependent variation. Methods for labelling samples for microarray experiments can be divided into two general categories: direct or indirect incorporation of modified nucleotides into cDNA or antisense RNA. All applications of cluster analysis in microarray research can be divided in two categories: clustering of genes by the similarity of their expression patterns across a number of different conditions or clustering of phenotypes by the similarity in gene-expression pattern.
