Reconstruction of Gene Regulatory Networks by Neuro-fuzzy Inference Systems

In this paper, we propose a new reconstruction method of gene regulatory networks (GRNs) from gene expression profiles obtained by DNA microarray experiments through a neuro-fuzzy inference system (NFIS). One of the major difficulties in reconstructing GRNs is caused by the noisy and uncertain information of gene expression profiles introduced during DNA microarray experiments or preprocessing of the raw data. In the proposed method, a gene expression profile is first transformed into a mapping form and then the transformed data are mapped into the NFIS. Finally, the resulting fuzzy rules are used to infer the relations. Since the relations are represented by fuzzy rules, the proposed method is robust to noisy and uncertain information. We illustrate the proposed method through a GRN represented by a linear model. It turns out that the NFIS is a useful framework for reconstruction of GRNs.