We have applied techniques of gene expression analysis to the analysis of human breast cancer by identifying metagene models with the capacity to discriminate breast tumors based on estrogen receptor (ER) status as well as the propensity for lymph node metastasis. We assess the utility and validity of these models in predicting status of tumors in cross-validation determinations. The practical value of such approaches relies on the ability not only to assess relative probabilities of clinical outcomes for future samples but also to provide an honest assessment of the uncertainties associated with such predictive classifications, based on the selection of gene subsets for each validation analysis. This latter point is of critical importance to the ability of applying these methodologies to clinical assessment of tumor phenotype. It is also clear from ER predictions that these analyses identify genes known to be involved in ER function but also identify new candidate genes involved in ER function. We believe these gene expression phenotypes have the potential to characterize the complex genetic alterations that typify the neoplastic state in a way that truly reflects the complexity of the regulatory pathways that are affected.