Discrete Logic Modelling Optimization to Contextualize Prior Knowledge Networks Using PRUNET

摘要: High-throughput technologies have led to the generation of an increasing amount data in different areas biology. Datasets capturing cell’s response its intra- and extra-cellular microenvironment allows such be incorporated as signed directed graphs or influence networks. These prior knowledge networks (PKNs) represent our current causality cellular signal transduction. New signalling is often examined interpreted conjunction with PKNs. However, biological contexts, cell type disease states, may distinct variants pathways, resulting misinterpretation new data. The identification inconsistencies between measured topologies, well training PKNs using context specific datasets (PKN contextualization), are necessary conditions construct reliable, predictive models, which challenges systems biology signalling. Here we present PRUNET, a user-friendly software tool designed address contextualization experimental conditions. As input, algorithm takes PKN expression profile two given stable steady states phenotypes. iteratively pruned evolutionary perform optimization process. This rests match predicted attractors discrete logic model (Boolean) Booleanized representation phenotypes, within population alternative subnetworks that evolves iteratively. We validated applying PRUNET four examples contextualized predict missing values simulate well-characterized perturbations. constitutes for automatic curation make it suitable describing processes under particular general applicability implemented makes variety processes, instance reprogramming transitions healthy states.