Automated extraction of information from biological literature promises to play an increasingly important role in knowledge discovery from high throughput approaches such as microarrays. We have developed an integrated system that combines protein/gene name dictionaries, synonymy dictionaries, natural language processing, and pattern matching rules to extract and organize gene relationships from full text articles. These articles were collected from 5 years (1999-2003) of 20 peer-reviewed journals in the field of molecular biology and biomedicine. As a pilot project, we examined the lists from a differential gene expression experiment examining the response of two mitogenic stimuli (EGF and S1P). Each gene was compared to the concepts extracted to discover co-regulatory patterns. This analysis resulted in a high correlation between genes in the common list and the biological processes of cell cycle and proliferation. This was to be expected since both EGF and S1P stimulate cell proliferation. The gene list unique to S1P stimulation resulted in a high correlation with the concept of cell stress response. This correlation was not immediately obvious from gene ontologies or pathway analysis. These data suggest that the knowledge bases resulting from full text and will assist in the understanding of gene lists generated from microarray experiments.