Living organisms are made of cells. Every cell contains a copy of the genome specific to that particular type of organism. The genome is a large molecule of DNA that encodes the structure of many thousands of proteins. In bacteria, at least 10% of these proteins are catalysts for a link in a network of so-called metabolic reactions, inter-converting chemicals termed metabolites. A typical bacterium will have a large metabolic network of over 1500 metabolic reactions interconverting 1000 metabolites. As a whole, a metabolic network is responsible for consuming high-energy chemicals from the environment, synthesizing precursors for macromolecule synthesis (including protein catalysts) and excreting low-energy chemicals back into the environment. A protein catalyst can be reutilized many thousands of times to catalyze a metabolic reaction before it ages and is degraded. Protein catalysts are themselves synthesized from metabolites, in a macromolecular synthesis network of reactions, but with reaction rates far slower than the rate of the metabolic reactions. Metabolic network function is complicated. Too complicated to be reasoned about descriptively without in advertently wandering into a logical contradiction. To aid understanding of complicated biochemical networks, the scientific field of molecular systems biology was born. At the core of molecular systems biology is the aim to construct, in a computational model, a complete representation of our understanding of a particular biochemical network under study. For metabolic networks, this requires the construction of a computational model of each metabolic reaction and the reactions necessary to synthesize each protein catalyst. Such models have reaction rates spread over many orders of magnitude: they are multiscale. Multiscale modeling of an integrated metabolic and macromolecular synthesis network is essential for representing our understanding of how these systems interact, for predicting consequences of this interaction and for predicting experiments that further our understanding of this interaction. Specialized software (based on numerical optimization algorithms) is required to process such integrated models. There is a need to improve the reliability of such software and thereby generate falsifiable predictions from the models in order to test if the model accurately represents the real-world biochemical network. Such models are developed for bacteria as they are comparatively simple compared to human biochemical networks. All biochemical networks share deep similarities, so development of generic computational tools for modeling one organism can very often be applied to model another similar organism. To understand a biochemical network via modeling is prerequisite to being able to control it. To control it is to be able to fix it when it is broken and to bring forth applications that stimulatethe biotechnology economy.

Public Health Relevance

Ultimately, the workings of the human body can only be fully understood when simulated in a computer, then compared with clinical data. This work aims to develop simulation software for understanding the workings of simple organisms, like bacteria.

National Institute of Health (NIH)
National Institute of General Medical Sciences (NIGMS)
Research Project--Cooperative Agreements (U01)
Project #
Application #
Study Section
Special Emphasis Panel (ZEB1-OSR-C (M3))
Program Officer
Lyster, Peter
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Stanford University
Engineering (All Types)
Schools of Engineering
United States
Zip Code
Yang, Laurence; Ma, Ding; Ebrahim, Ali et al. (2016) solveME: fast and reliable solution of nonlinear ME models. BMC Bioinformatics 17:391
Utrilla, Jose; O'Brien, Edward J; Chen, Ke et al. (2016) Global Rebalancing of Cellular Resources by Pleiotropic Point Mutations Illustrates a Multi-scale Mechanism of Adaptive Evolution. Cell Syst 2:260-71
Fleming, Ronan M T; Vlassis, Nikos; Thiele, Ines et al. (2016) Conditions for duality between fluxes and concentrations in biochemical networks. J Theor Biol 409:1-10
Yang, Laurence; Tan, Justin; O'Brien, Edward J et al. (2015) Systems biology definition of the core proteome of metabolism and expression is consistent with high-throughput data. Proc Natl Acad Sci U S A 112:10810-5
Ebrahim, Ali; Almaas, Eivind; Bauer, Eugen et al. (2015) Do genome-scale models need exact solvers or clearer standards? Mol Syst Biol 11:831
Latif, Haythem; Szubin, Richard; Tan, Justin et al. (2015) A streamlined ribosome profiling protocol for the characterization of microorganisms. Biotechniques 58:329-32
O'Brien, Edward J; Palsson, Bernhard O (2015) Computing the functional proteome: recent progress and future prospects for genome-scale models. Curr Opin Biotechnol 34:125-34
Choi, Sou-Cheng T; Saunders, Michael A (2014) Algorithm 937: MINRES-QLP for Symmetric and Hermitian Linear Equations and Least-Squares Problems. ACM Trans Math Softw 40:
Meng, Xiangrui; Saunders, Michael A; Mahoney, Michael W (2014) LSRN: A PARALLEL ITERATIVE SOLVER FOR STRONGLY OVER- OR UNDERDETERMINED SYSTEMS. SIAM J Sci Comput 36:C95-C118
Liu, Joanne K; O'Brien, Edward J; Lerman, Joshua A et al. (2014) Reconstruction and modeling protein translocation and compartmentalization in Escherichia coli at the genome-scale. BMC Syst Biol 8:110

Showing the most recent 10 out of 26 publications