Fuzzy Model for Gene Regulatory Network

Abstract

Gene regulatory networks influence development and evolution in living organism. The advent of microarray technology has challenged computer scientists to develop better algorithms for modeling the underlying regulatory relationship in between the genes. Recently, a fuzzy logic model has been proposed to search microarray datasets for activator/repressor regulatory relationship. We improve this model for searching regulatory triplets by means of predicting changes in expression level of the target over interval time points based on input expression level, and comparing them with actual changes. This method eliminates possible false predictions from the classical fuzzy model thereby allowing a wider search space for inferring regulatory relationship. We also introduce a novel pre-processing technique using fuzzy logic that can group genes having similar changes in expression profile over all available intervals in the microarray data. This technique eliminates redundant computation performed by the proposed model. Saccharomyces cerevisiae data was applied to the model and 548 activator/repressor regulatory triplets were inferred from the data. These improvements will increase feasibility of using fuzzy logic for understanding the relationship between genes using microarray technology.