Pacific Northwest National Laboratory and University of California, Riverside
Expertise and Interests: Our research seeks to understand how cells change as a function of the environmental perturbations, whether the environment is soil, the human gut or a biofuel reactor. On a more ambitious scale, our research seeks to understand the principles of self-organization, especially with regard to scaling in space and time and the emergence of biological function and mechanisms.
Simulation and Modeling: We have developed advanced simulation technology that will bring much higher predictive modeling to biology. Briefly, statistical thermodynamics can be used to reformulate the law of mass action so that biological processes can be modeled more faithfully and allow for the inference of rate constants from quantitative metabolomics data. This approach goes beyond the modeling of flux using constraint-based methods and provides information on pathway thermodynamics, power characteristics of reactions and prediction of metabolite concentrations, rate constants and especially post-translational regulation. This technology allows us to quantitatively model the energetics and dynamics of metabolism and cellular processes.
Data Analysis and Modeling: Our ability to effectively utilize data is often limited due to inherent noise in the data, lack of statistical confidence and the use of ad hoc or black box models for integrating and making biological sense of the data. Models representing domain knowledge and scientific principles, however, can provide the prior information needed to interpret large-scale proteomics, metabolomics and gene expression data. Our interests are in combining statistical thermodynamics, information theory, control theory and machine learning methods to analyze the data using models as prior knowledge. Reinforcement learning exploiting physics-based Markov models is one way to accomplish this.
IMAG project: Multi-scale Model of Circadian Rhythms