Amos Ben-Zvi

Metabolomic profiling of asthma: diagnostic utility of urine nuclear magnetic resonance spectroscopy.

BACKGROUND The ability to diagnose and monitor asthma on the basis of noninvasive measurements of airway cellular dysfunction is difficult in the typical clinical setting. OBJECTIVE Metabolomics is the study of molecules created by cellular metabolic pathways. We hypothesized that the metabolic activity of children with asthma would differ from healthy children without asthma. Furthermore, children having an asthma exacerbation would be different compared with children with stable asthma in outpatient clinics. Finally, we hypothesized that (1)H-nuclear magnetic resonance (NMR) would measure such differences using urine samples, one of the least invasive forms of biofluid sampling. METHODS Children (135 total, ages 4-16 years) were enrolled, having met the criteria of healthy controls (C), stable asthma in the outpatient clinic (AO), or unstable asthma in the emergency department (AED). Partial least squares discriminant analysis was performed on the NMR data to create models of separation (70 metabolites were measured/urine sample). Some NMR data were withheld from modeling to be run blindly to determine possible diagnostic accuracy. RESULTS On the basis of the model of AO versus C, 31 of 33 AO samples were correctly diagnosed with asthma (94% accuracy). Only 1 of 20 C samples was incorrectly labeled as asthma (5% misclassification). On the basis of the AO versus AED model, 31 of the 33 AO samples were correctly diagnosed as outpatient asthma (94% accurate). CONCLUSION This is the first report suggesting that (1)H-NMR analysis of human urine samples has the potential to be a useful clinical tool for physicians treating asthma.

The dynamics of heterotrophic algal cultures.

In this work, the time varying characteristics of microalgal cultures are investigated. Microalgae are a promising source of biofuels and other valuable chemicals; a better understanding of their dynamic behavior is, however, required to facilitate process scale-up, optimization and control. Growth and oil production rates are evaluated as a function of carbon and nitrogen sources concentration. It is found that nitrogen has a major role in controlling the productivity of microalgae. Moreover, it is shown that there exists a nitrogen source concentration at which biomass and oil production can be maximized. A mathematical model that describes the effect of nitrogen and carbon source on growth and oil production is proposed. The model considers the uncoupling between nutrient uptake and growth, a characteristic of algal cells. Validity of the proposed model is tested on fed-batch cultures.

Model-Based Therapeutic Correction of Hypothalamic-Pituitary-Adrenal Axis Dysfunction

The hypothalamic-pituitary-adrenal (HPA) axis is a major system maintaining body homeostasis by regulating the neuroendocrine and sympathetic nervous systems as well modulating immune function. Recent work has shown that the complex dynamics of this system accommodate several stable steady states, one of which corresponds to the hypocortisol state observed in patients with chronic fatigue syndrome (CFS). At present these dynamics are not formally considered in the development of treatment strategies. Here we use model-based predictive control (MPC) methodology to estimate robust treatment courses for displacing the HPA axis from an abnormal hypocortisol steady state back to a healthy cortisol level. This approach was applied to a recent model of HPA axis dynamics incorporating glucocorticoid receptor kinetics. A candidate treatment that displays robust properties in the face of significant biological variability and measurement uncertainty requires that cortisol be further suppressed for a short period until adrenocorticotropic hormone levels exceed 30% of baseline. Treatment may then be discontinued, and the HPA axis will naturally progress to a stable attractor defined by normal hormone levels. Suppression of biologically available cortisol may be achieved through the use of binding proteins such as CBG and certain metabolizing enzymes, thus offering possible avenues for deployment in a clinical setting. Treatment strategies can therefore be designed that maximally exploit system dynamics to provide a robust response to treatment and ensure a positive outcome over a wide range of conditions. Perhaps most importantly, a treatment course involving further reduction in cortisol, even transient, is quite counterintuitive and challenges the conventional strategy of supplementing cortisol levels, an approach based on steady-state reasoning.

Optimization of microalgal productivity using an adaptive, non-linear model based strategy

The optimization of biomass and oil productivities in heterotrophic cultures of Auxenochlorella protothecoides was achieved using a non-linear model-based approach. A 10-fold increase in the average biomass productivity, and a 16-fold increase in the maximum productivity, was observed with respect to batch cultures as a result of the proposed optimization strategy. Final cell density in the optimized culture was 144 g/L (dry weight), with 49.4%w/w oil content. Maximum lipid productivity was 20.16 g/L d, achieved during the exponential growth phase at an average cell density of 86 g/L. Lipid productivity in the optimized microalgal culture was higher than previously reported values for other oleaginous microorganisms. Oil composition analysis showed that the oil has a high quality as biodiesel precursor. The higher productivity and excellent lipid profile of the optimized microalgal culture make A. protothecoides an exceptional source for biodiesel production and a potential source of single cell oil for other applications.

Receding Horizon Experiment Design with Application in SOFC Parameter Estimation

Abstract In this work the problem of optimal input design (OID) in a receding-horizon framework for online parameter estimation is solved. The designed optimum input is used for dynamic experiment and subsequent estimation of parameters. A fuel cell experiment design and parameter estimation problem is investigated through the proposed approach. Some of the issues related to the application of the proposed method are examined and guidelines for selecting appropriate experimental settings are provided.

Asymptotic parameter sensitivity for polynomial dynamical systems

The effect of parameter perturbations on the trajectories of dynamical systems has been extensively studied in the literature. The effect of parameter values on the number and type of rest-points in a dynamical system has received less attention. For systems that exhibit input multiplicities, the sensitivity of system trajectories to parameter values can be discontinuous when state trajectories are near the boundary between the basins of different attractors. In this work, a computational scheme for conducting a single and joint parameter sensitivity analysis for systems with multiple steady states is presented. The proposed approach is computationally efficient and relies on algebraic geometric tools to obtain a simplified polynomial representation of the system at steady state. The proposed approach is applied to a model of the human hypothalamic-pituitary-adrenal axis and is shown to be relevant to the development of mechanistic models of chronic diseases.

Challenges in Reverse Engineering of Gene Networks from Algebraic Perspectives

Abstract Reverse engineering of complex gene networks is a significant step towards understanding various biological processes. In this work, a novel algorithm for reverse engineering gene networks on a genome-wide scale using a noisy gene expression data is proposed. Under the proposed scheme, the challenges in reverse engineering of gene networks from these gene expression datasets are highlighted. Also, the parameter space describing gene interaction is partitioned into estimable and inestimable linear subspaces. The estimable linear subspace is obtained by using principal components analysis (PCA), the Akaike information criterion (AIC), and jackknifing. Furthermore, the approach is tested and validated using a simulated gene network model.

NONLINEAR DYNAMICAL ANALYSIS AND OPTIMIZATION FOR BIOLOGICAL/ BIOMEDICAL SYSTEMS

As mathematical models are increasingly available for biological/biomedical systems, dynamic optimization can be a useful tool for manipulating systems. Dynamic optimization is a computational tool for finding a sequence of optimal actions to attain desired outcomes from the system. This chapter discusses two dynamic optimization algorithms, model predictive control and dynamic programming, in the context of finding optimal treatment strategy for correcting hypothalamic-pituitary-adrenal (HPA) axis dysfunction. It is shown that dynamic programming approach has the advantage over the model predictive control (MPC) methodology in terms of robustness to error in parameter estimates and flexibility of accommodating clinically relevant objective function.

An MPC-Based Approach to Robust Control of a Hypothalamic-Pituitary-Adrenal Axis Model

Abstract The hypothalamic-pituitary-adrenal (HPA) axis is a neuroendocrine control system regulating the stress response of the human body as well as affecting mood and cognition. HPA dysfunction has been associated with a variety of disorders including Fibromyalgia, and chronic fatigue syndrome (CFS). Recently, a lumped parameter model of the HPA axis which includes the effect of pharmacological agents has been proposed. In this work, an analysis of the HPA axis dynamics near rest points is used to develop an approach for robust control of the HPA system.

ON THE EFFECT OF UN-IDENTIFIABILITY ON CONTROL

Abstract Mathematical models often depend on unknown parameters that must be identified. Un-identifiable models have parameters that cannot be identified from input-output data. In this work, it is shown that un-identifiability can affect the closed-loop performance of systems. This conclusion holds even for minimal systems. It is shown that a change of coordinates can be used to transform any linear, time-invariant un-identifiable system into one that is identifiable up to a change in initial conditions. For such systems, it is possible to construct controller/observer pairs that do not depend on any un-identifiable parameters.