An ensemble approach for gene regulatory network study in rice blast

Abstract

Gene Regulatory Network (GRN) is important to understand complex biological process like disease. An ensemble approach for network construction leads to more accurate and consistent GRN than individual methods. In this study we have combined the networks obtained from correlation, partial least square, principal component regression and ridge regression techniques using Fisher’s weighted method. The network was ensemble considering the significant edges selected from individual methods. The method was applied to gene expression data of rice blast disease to construct GRN. The stability of the obtained ensemble network was higher than the networks obtained from individual methods.