ABSTRACT
In 1999, the Science paper by Golub et al. (1999) started a new era in the world of clinical diagnostics and prognosis by the introduction of high throughput data measured through the use of microarrays. These devices enable measuring tens of thousands of variables in one single experiment. Originally, micro-arrays were mainly used to measure gene expression levels in tissue or serum, but presently they are also used for measuring DNA content and DNA methylation. In the meantime new technology already has entered in the form of the so-called proteomic data produced by mass-spectrometry based techniques like MALDI-TOF. The pioneering paper breaking the ground for clinical applications of proteomics is the one by Petricoin et al. (2002). The focus in this part of the book will be on gene expression data, but the statistical methodology carries over to all type of high-dimensional data.
