Data and model driven hybrid approach to activity scoring of cyclic pathways

Analysis of large scale -omics data based on a single tool remains inefficient to reveal molecular basis of cellular events. Therefore, data integration from multiple heterogeneous sources is highly desirable and required. In this study, we developed a data- and model-driven hybrid approach to evaluate biological activity of cellular processes. Biological pathway models were taken as graphs and gene scores were transferred through neighbouring nodes of these graphs. An activity score describes the behaviour of a specific biological process was computed by owing of converged gene scores until reaching a target process. Biological pathway model based approach that we describe in this study is a novel approach in which converged scores are calculated for the cellular processes of a cyclic pathway. The convergence of the activity scores for cyclic graphs were demonstrated on the KEGG pathways. © 2011 Springer Science+Business Media B.V.