Development of network inference among LexA/ RecA-dependent manner genes in the SOS response

Abstract

The expression of many genes varies under different conditions. The relationships among genes can be identified from DNA microarray data. To better understand the SOS response, I developed a network model of whole LexA/RecA-dependent manner genes in two genotypes of Escherichia coli from expression data after ultraviolet (UV) irradiation. First, all LexA/RecA-dependent manner genes were estimated by a combination of the Mann-Whitney U test and hierarchical clustering. Second, the relationships between genes were inferred from a graphical Gaussian model (GGM) combined with hierarchical clustering. Here, I incorporate a step-by-step procedure of the method developed in my previous study to identify interactions within larger clusters. The analysis suggested the presence of a sequential relationship between LexA/RecA-dependent manner gene groups in which the association between neighboring groups was positive while that between non-neighboring groups was negative. I suggest the inferred network provides information on the functional role of poorly defined genes whose expression patterns change according to LexA/RecA. Although relationships in the network require biological validation, the model presented here provides insights into potential LexA/RecA regulation following UV irradiation.