Many researchers, such as those at the National Center for Toxicogenomics (NCT) and at the National Toxicology Program (NTP),  are conducting time course and dose response studies to understand the pattern of gene expression in response to dose of a compound or duration of exposure to a compound at a given dose.

In this talk we describe ORIOGEN, an algorithm for selecting and  clustering genes according to their time-course or dose-response profiles using data from microarray experiments. ORIOGEN is based on the order restricted inference methodology developed in Hwang and Peddada (1994, Annals of Statistics),  in Peddada et al. (2001, Biometrics) and Peddada et al. (2003). 

We describe the methodology for time-course experiments although it is applicable to any ordered set of treatments.  Candidate temporal profiles are defined in terms of inequalities between the mean gene expression levels at the time points.  ORIOGEN  selects genes when they meet a bootstrap-based criterion for statistical significance and assigns each selected gene to  the best fitting candidate profile. We illustrate the methodology  using data from a cDNA microarray experiment  in which a breast cancer cell line was stimulated with estrogen for different time intervals. In this example, our method was able to identify several biologically interesting genes that previous analyses failed to reveal.