Recent advances in microarray technology have enabled the measurement of expression of mRNA transcripts for thousands of genes simultaneously. In addition to being a significant advance in high throughput screening, this technology redefines how biologists are able to view biochemical events during the development of a pathological state and intervention by therapeutic agents. However, many statistical challenges remain to analyze these large data sets (up to 13.000 genes per chip) to minimize false positive rates, and to develop appropriate statistical tools to determine relationships between the expression of large sets of genes. We have recently used this technology to determine changes in gene expression in HIV infected cells, and the effect of a new anti-HIV compound on this expression. The presentation will illustrate some of the statistical tools currently available to identify gene targets, and, highlight some tools that need to be developed to mathematically describe how biochemical networks that are activated or deactivated during normal and pathological functioning.