Paul Barton, Ph.D.
Department of Chemical Engineering
Professor of Chemical Engineering
Ph.D. Chemical Engineering, 1992
Imperial College, London, UK
The process systems engineering laboratory at MIT is conducting a broad program of research related to modeling, simulation, optimization and design of large-scale dynamic systems. Applications are drawn both from traditional chemical process industries and from less traditional areas such as pharmaceutical and biochemical processes, micro-process systems, signaling and regulation in biological systems, complex chemical reaction mechanisms such as those in combustion systems, and self-assembly of nanostructured materials. We are engaged in two interdisciplinary collaborations to apply modeling tools and approaches to problems in biology and biochemistry.
Several CSBi laboratories are using the ABACUSS software platform, which we developed to aid in modeling and simulating complex processes. ABACUSS provides a high-level environment with a graphical interface that lets users formulate, debug and numerically solve large-scale dynamic models of the kinetics of biochemical networks. ABACUSS currently supports the simulation, sensitivity analysis and parameter estimation of large-scale discrete/continuous models based on differential-algebraic equations. Currently, we are incorporating into ABACUSS new algorithms that we developed for mixed-integer and dynamic optimization.
Dynamic optimization of biochemical networks
Biological systems have evolved so that they can undertake sophisticated decisions and control complex processes while remaining resilient and flexible. For example, signal transduction cascades provide an important set of pathways that sense and process extracellular signals and trigger internal cellular events. We are collaborating with Bruce Tidors laboratory to develop modeling tools to analyze and compare biological networks and classes of networks. Our hypothesis is that because biological networks have evolved to function well in a complex environment, we expect that they are at or near their optimal state among a class of related networks. Therefore, the key to understanding enzyme networks is to elucidate the precise sense in which they are optimal and the class of related networks with respect to which they are optimal. This can be accomplished by posing and solving an optimization problem and comparing the solution with networks that can be observed in nature. We are using this approach with two model systems: mitogen-activated protein (MAP) kinase, an enzyme superfamily that is involved in a variety of signaling pathways; and synthetic genetic networks, a well-defined experimental system composed of DNA regulatory elements that can be used to re-create signal processing units involving feedback control and other mechanisms.
Bayesian analysis of cell migration
Our collaboration with Douglas Lauffenburgers laboratory focuses on cell migration, which is a key process in inflammation, wound healing, embryogenesis and tumor cell metastasis. Small peptide ligands such as epidermal growth factor (EGF) and fibronectin bind at surface receptors and can trigger a complex signaling cascade within the cell that may alter cell mobility or trigger cell division. We are developing better statistical methods that will allow us to understand and model how signaling molecules and other environmental stimuli affect macroscopic parameters of cell behavior such as movement and turning as a function of time. We are using Bayesian techniques to address the biggest challenge in this work, which is the development of good computational models that match the experimental system and can be used for experimental design and parameter estimation.
- "Global Optimization of Linear Hybrid Systems with Explicit Transitions", In press, Systems & Control Letters, August 2003, (with C. K. Lee and A. B. Singer).
- "Global Solution of Linear Dynamic Embedded Optimization Problems", In press, Journal of Optimization Theory and Applications, June 2003, (with A.B. Singer).
- "Modeling, Simulation, Sensitivity Analysis and Optimization of Hybrid Systems", ACM Transactions on Modeling and Computer Simulation, 12(4):256-289, 2002, (with C. K. Lee). "Hidden Discontinuities and Parametric Sensitivity Analysis", SIAM Journal on Scientific Computing, 23(6):1862-1875, (2002), (with J. E. Tolsma).
- "On Upgrading the Numerics in Combustion Chemistry Codes", Combustion and Flame, 128(3):270-291, (2002), (with D. A. Schwer, J. E. Tolsma and W. H. Green, Jr.).
Last Updated: April 15, 2008