Effective Identification of Negative Regulation Patterns of Protein Kinases

Abstract

Recent studies point to the fact that protein kinases play an important role in the regulation of cellular pathways and show great potential in disease treatment. Thus, it is critical to discover characterized regulatory patterns of protein kinases in signaling pathway. There have been considerable efforts to explore the activities of protein kinases. However, the study of negative regulation patterns has been largely overlooked and undeveloped. This paper aims to identify inhibitory regulatory correlations of protein kinase according to negative association rule mining. Especially, mutual information is applied to sort out the items with strong dependency and the minimum support threshold is computed by support constraints to control rule generation. The obtained rules not only reveal the relationships between subunits of protein kinases and between subunits and stimuli, but also provide novel pharmacological insight into drug design for diseases.