Biomedical Engineering Reference
In-Depth Information
2
GENE SELECTION STRATEGIES IN
MICROARRAY EXPRESSION
DATA: APPLICATIONS TO
CASE-CONTROL STUDIES
Gustavo A. Stolovitzky
IBM Computational Biology Center, Yorktown Heights, New York
Over the last decade we have witnessed the rise of the gene expression array assay as a
new experimental paradigm to study the cellular state at the whole genome scale. This
technology has allowed considerable progress in the identification of markers associated
with human disease mechanisms, and in the molecular characterization of diseases such
as cancer, by careful characterization of genes involved directly or indirectly in the dis-
ease. A typical gene expression experiment provides scientists with an enormous amount
of data. Analysis of these data, and interpretation of the ensuing results, have attracted the
attention of many researchers, who have developed new ways of interrogating the expres-
sion data. In this chapter we will review some of these recent efforts, emphasizing the
need to make use of batteries of methods rather than one method in particular, as well as
the need to properly validate results with independent data sets. The application of DNA
array technology for use in disease diagnostics will be exemplified in the case of chronic
lymphocytic leukemia.
1.
INTRODUCTION
DNA microarrays constitute one of the most powerful high-throughput
technologies in molecular biology today. It has emerged in recent years as a
powerful tool that provides a glimpse into the complexities of cellular behavior
Address correspondence to: Gustavo A. Stolovitzky, IBM Computational Biology Center, 1101
Kitchawan Road, Yorktown Heights, NY 10598 (gustavo@us.ibm.com).
679
