From data towards knowledge: Revealing the architecture of signaling systems by unifying knowledge mining and data mining of systematic perturbation data

摘要

Genetic and pharmacological perturbation experiments, such as deleting a geneand monitoring gene expression responses, are powerful tools for studyingcellular signal transduction pathways. However, it remains a challenge toautomatically derive knowledge of a cellular signaling system at a conceptuallevel from systematic perturbation-response data. In this study, we explored aframework that unifies knowledge mining and data mining approaches towards thegoal. The framework consists of the following automated processes: 1) applyingan ontology-driven knowledge mining approach to identify functional modulesamong the genes responding to a perturbation in order to reveal potentialsignals affected by the perturbation; 2) applying a graph-based data miningapproach to search for perturbations that affect a common signal with respectto a functional module, and 3) revealing the architecture of a signaling systemorganize signaling units into a hierarchy based on their relationships.Applying this framework to a compendium of yeast perturbation-response data, wehave successfully recovered many well-known signal transduction pathways; inaddition, our analysis have led to many hypotheses regarding the yeast signaltransduction system; finally, our analysis automatically organized perturbedgenes as a graph reflecting the architect of the yeast signaling system.Importantly, this framework transformed molecular findings from a gene level toa conceptual level, which readily can be translated into computable knowledgein the form of rules regarding the yeast signaling system, such as "if genesinvolved in MAPK signaling are perturbed, genes involved in pheromone responseswill be differentially expressed".