Network Descriptors to Capture the Dynamic Changes of Networks in Temporal Microarray Data Sets

Abstract

The analysis of temporal microarray datasets usually focused on the expression patterns of genes, and genes with similar expression patterns are clustered. However, the relationships among the clustered genes are not clear except for the plausible co-regulated connections. Also, there are no established methods to quantify the changes among the data sets in different stages. We have combined protein-protein interaction data with micorarray expression profiles to reveal the possible interactions among genes with/without similar expression patterns. Based on the features of networks, we propose a quantitative measure to evaluate the changes between two microarray datasets in different conditions or time points. The quantified scores could in turn be used as a criterion to identify significant sub-networks, which may be used to establish the functional connections to the specified conditions or factors.