A Discrete Event Based Stochastic Simulation Approach For Studying The Dynamics Of Biological Networks
MetadataShow full item record
With increasing availability of data resources on the molecular parts of a living cell, biologists are focusing on holistic understanding of cellular mechanisms and the emergent dynamics arising out of their complex interactions. Comprehending the fine-grained signal specificity, gene regulation and feedback mechanisms of molecular interactions at a network level forms a central theme of systems biology. With the speed and sophistication of computational methods, in silico modeling and simulation techniques have become a powerful tool for biologists challenged with understanding the system complexity of biological networks. Numerical simulation of classical chemical kinetics (CCK), agent-based simulations of biological processes, and linear optimization models of metabolic networks, have been applied to the study of cellular behaviors with varying degrees of success. The spatio-temporal scales of cellular processes, coupled with the knowledge gap and complexity of biological networks limit the application of existing computational techniques. In this thesis, we present a network-centric modeling and simulation approach to systematically study the stochastic dynamics of cellular processes at a molecular level. The central theme of our approach revolves around abstracting a complex biological process as a collection of discrete, interacting molecular entities driven in time by a set of discrete molecular events (bioEvents). We develop the discrete-event based simulation engine, called iSimBioSys, together with an integrated database of biological pathways, which captures the temporal dynamics of the molecular entities through stochastic interactions of different bioEvents. With an illustrative case study of signal transduction networks in bacterial cells, we highlight the efficiency of a discrete event based approach in capturing high-level system dynamics of a biological process, particularly in reproducing the switching effect of the PhoPQ pathway in Salmonella cells as reported in experimental work. Next, we build a detailed stochastic model for the fundamental process of gene expression in prokaryotic cells and study the molecular events of transcription and translation using the proposed simulation framework. Our results identify the role of transcriptional and translation machinery in controlling the burstiness of protein generation. We extend our simulator to incorporate a hybrid algorithm which combines stochastic models of signaling and regulatory events with a flow-based model for metabolic networks. In order to validate the efficacy of the hybrid simulation approach, we develop an integrated database of signaling and metabolic networks in the bacterial cell Escherichia Coli. The hybrid simulation recreates experimentally observed regulation of metabolic flux distributions in the network while providing new insights into the mechanism of regulation.