Models Of Reaction Networks Research Articles

Methods of Model Reduction for Large-Scale Biological Systems: A Survey of Current Methods and Trends

Complex models of biochemical reaction systems have become increasingly common in the systems biology literature. The complexity of such models can present a number of obstacles for their practical use, often making problems difficult to intuit or computationally intractable. Methods of model reduction can be employed to alleviate the issue of complexity by seeking to eliminate those portions of a reaction network that have little or no effect upon the outcomes of interest, hence yielding simplified systems that retain an accurate predictive capacity. This review paper seeks to provide a brief overview of a range of such methods and their application in the context of biochemical reaction network models. To achieve this, we provide a brief mathematical account of the main methods including timescale exploitation approaches, reduction via sensitivity analysis, optimisation methods, lumping, and singular value decomposition-based approaches. Methods are reviewed in the context of large-scale systems biology type models, and future areas of research are briefly discussed.

In-situ measurement of residence time distributions in a turbulent oxy-fuel gas-flame combustor

For improving the design of combustors, the knowledge of residence-time distributions (RTD) is important as they influence exhaust gas compositions. Measuring RTDs in combustors is challenging, due to high temperatures, chemical reactions, the presence of particles or corrosive species as well as high turbulence levels. This paper presents a technique for the in situ measurement of RTDs in combustors. Based on tunable diode laser absorption spectroscopy (TDLAS), the temporal evolution of the concentration of tracers is tracked simultaneously at the combustion chamber inlet and outlet. Using either air or mixtures of oxygen and carbon dioxide (oxy-fuel atmosphere) as oxidants, the method is applied to reacting and non-reacting operating conditions in a 20-kWth methane combustor. For reacting conditions, hydrogen chloride is used as a tracer, whereas for non-reacting conditions methane was chosen. Depending on the tracer, for a repetition rate of approximately 2 kHz detection limits of 16–40 ppmV are achieved. For deriving RTDs, low-pass filtering is compared to reactor network modeling. Different RTDs observed for varying operating conditions are discussed.

A general reaction network and kinetic model of the hydrothermal liquefaction of microalgae Tetraselmis sp.

In this work, the hydrothermal liquefaction (HTL) of microalgal Tetraselmis sp. was conducted at various reaction temperatures (250–350°C) and reaction times (10–60min). A general reaction network and a quantitative kinetic model were proposed for the HTL of microalgae. In this reaction network, the primary decomposition of lipids, proteins, and carbohydrates generated heavy oil (HO), light oil (LO), and aqueous-phase (AP) products. Then, reversible interconversions and further decomposition of these product fractions to produce gas product were followed. The model accurately captures the trends observed in the experimental data. Analyses of the kinetic parameters (reaction rate constants and activation energies) suggested the dominant reaction pathways as well as the contribution of the biochemical compositions to the bio-oil yield. Finally, the kinetic parameters calculated from the model were utilized to explore the parameter space in order to predict the liquefaction product yields depending on the reaction time and temperature.

A new framework for modeling coal devolatilization and combustion in boiler furnaces

This paper presents a new framework for the modeling of coal-fired boiler furnaces. The input required for the model is the ultimate analysis of a coal sample. The model accounts for devolatilization followed by gas-phase combustion. The devolatilization model used in this work is taken from published literature with slight modifications to match the numerical predictions with experimental measurements. This work also develops a reactor network model for simulating the performance of boiler furnaces. For the seamless integration of kinetic models of coal devolatilization and combustion with furnace numerical model, the thermochemistry data of several hypothetical and intermediate species involved in devolatilization chemistry are evaluated in the form of 14 coefficient National Aeronautics and Space Administration polynomials. The capability of the model for predicting the furnace temperature and product composition is demonstrated by simulating a single-zone model. Copyright © 2017 John Wiley & Sons, Ltd.

Sensitivity estimation for stochastic models of biochemical reaction networks in the presence of extrinsic variability.

Determining the sensitivity of certain system states or outputs to variations in parameters facilitates our understanding of the inner working of that system and is an essential design tool for the de novo construction of robust systems. In cell biology, the output of interest is often the response of a certain reaction network to some input (e.g., stressors or nutrients) and one aims to quantify the sensitivity of this response in the presence of parameter heterogeneity. We argue that for such applications, parametric sensitivities in their standard form do not paint a complete picture of a system's robustness since one assumes that all cells in the population have the same parameters and are perturbed in the same way. Here, we consider stochasticreaction networks in which the parameters are randomly distributed over the population and propose a new sensitivity index that captures the robustness of system outputs upon changes in the characteristics of the parameter distribution, rather than the parameters themselves. Subsequently, we make use of Girsanov's likelihood ratio method to construct a Monte Carlo estimator of this sensitivity index. However, it turns out that this estimator has an exceedingly large variance. To overcome this problem, we propose a novel estimation algorithm that makes use of a marginalization of the path distribution of stochasticreaction networks and leads to Rao-Blackwellized estimators with reduced variance.

Multiscale Reactor Network Simulation of an Entrained Flow Biomass Gasifier: Model Description and Validation

AbstractThis paper describes the development of a multiscale equivalent reactor network model for pressurized entrained flow biomass gasification to quantify the effect of operational parameters on the gasification process, including carbon conversion, cold gas efficiency, and syngas methane content. The model, implemented in the commercial software Aspen Plus, includes chemical kinetics as well as heat and mass transfer. Characteristic aspects of the model are the multiscale effect caused by the combination of transport phenomena at particle scale during heating, pyrolysis, and char burnout, as well as the effect of macroscopic gas flow, including gas recirculation. A validation using experimental data from a pilot‐scale process shows that the model can provide accurate estimations of carbon conversion, concentrations of main syngas components, and cold gas efficiency over a wide range of oxygen‐to‐biomass ratios and reactor loads. The syngas methane content was most difficult to estimate accurately owing to the unavailability of accurate kinetic parameters for steam methane reforming.

Steam‐air blown bubbling fluidized bed biomass gasification (BFBBG): Multi‐scale models and experimental validation

During fluidized bed biomass gasification, complex gas‐solid mixing patterns and numerous chemical and physical phenomena make identification of optimal operating conditions challenging. In this work, a parametric experimental campaign was carried out alongside the development of a coupled reactor network model which successfully integrates the individually validated sub‐models to predict steady‐state reactor performance metrics and outputs. The experiments utilized an integrated gasification system consisting of an externally‐heated, bench‐scale, 4‐in., 5 kWth, fluidized bed steam/air blown gasifier fed with woody biomass equipped with a molecular beam mass spectrometer to directly measure tar species. The operating temperature (750–850°C) and air/fuel equivalence ratio (ER = 0–0.157) were independently varied to isolate their effects. Elevating temperature is shown to improve the char gasification rate and reduce tar concentrations. Air strongly impacts the composition of tar, accelerating the conversion of lighter polycyclic‐aromatic hydrocarbons into soot precursors, while also improving the overall carbon conversion. © 2016 American Institute of Chemical Engineers AIChE J, 63: 1543–1565, 2017

A Bayesian Learning Approach to Modeling Pseudoreaction Networks for Complex Reacting Systems: Application to the Mild Visbreaking of Bitumen

A data-mining and Bayesian learning approach is used to model the reaction network of a low-temperature (150–400 °C) visbreaking process for field upgrading of oil sands bitumen. Obtaining mechanistic and kinetic descriptions for the chemistry involved in this process is a significant challenge because of the compositional complexity of bitumen and the associated analytical challenges. Lumped models based on a preconceived reaction network might be unsatisfactory in describing the key conversion steps of the actual process. Fourier transform infrared spectra of products produced at different operating conditions (temperature and time of processing) of the visbreaking process were collected. Bayesian agglomerative hierarchical cluster analysis was employed to obtain groups of pseudospecies with similar spectroscopic properties. Then, a Bayesian structure-learning algorithm was used to develop the corresponding reaction network. The final reaction network model was compared to the anticipated reaction netwo...

A combined model reduction algorithm for controlled biochemical systems

BackgroundSystems Biology continues to produce increasingly large models of complex biochemical reaction networks. In applications requiring, for example, parameter estimation, the use of agent-based modelling approaches, or real-time simulation, this growing model complexity can present a significant hurdle. Often, however, not all portions of a model are of equal interest in a given setting. In such situations methods of model reduction offer one possible approach for addressing the issue of complexity by seeking to eliminate those portions of a pathway that can be shown to have the least effect upon the properties of interest.MethodsIn this paper a model reduction algorithm bringing together the complementary aspects of proper lumping and empirical balanced truncation is presented. Additional contributions include the development of a criterion for the selection of state-variable elimination via conservation analysis and use of an ‘averaged’ lumping inverse. This combined algorithm is highly automatable and of particular applicability in the context of ‘controlled’ biochemical networks.ResultsThe algorithm is demonstrated here via application to two examples; an 11 dimensional model of bacterial chemotaxis in Escherichia coli and a 99 dimensional model of extracellular regulatory kinase activation (ERK) mediated via the epidermal growth factor (EGF) and nerve growth factor (NGF) receptor pathways. In the case of the chemotaxis model the algorithm was able to reduce the model to 2 state-variables producing a maximal relative error between the dynamics of the original and reduced models of only 2.8% whilst yielding a 26 fold speed up in simulation time. For the ERK activation model the algorithm was able to reduce the system to 7 state-variables, incurring a maximal relative error of 4.8%, and producing an approximately 10 fold speed up in the rate of simulation. Indices of controllability and observability are additionally developed and demonstrated throughout the paper. These provide insight into the relative importance of individual reactants in mediating a biochemical system’s input-output response even for highly complex networks.ConclusionsThrough application, this paper demonstrates that combined model reduction methods can produce a significant simplification of complex Systems Biology models whilst retaining a high degree of predictive accuracy. In particular, it is shown that by combining the methods of proper lumping and empirical balanced truncation it is often possible to produce more accurate reductions than can be obtained by the use of either method in isolation.

Numerical study assessing various ammonia/methane reaction models for use under gas turbine conditions

Ammonia as an alternative fuel and hydrogen carrier has received increased attention in recent years. To explore the potential of co-firing ammonia with methane for power generation, studies involving robust mathematical analyses are required to progress towards industrial implementation. To explore a best suited mechanism for ammonia/methane combustion in gas turbines, five different detailed mechanisms were compared to identify their accuracy to represent the reaction kinetics under real gas turbine combustor conditions. Ignition delay time was compared with recent literature showing that the mechanisms of Tian and Teresa exhibit the best accuracy over a large range of conditions. A 1D simulation was also conducted using a Chemical Reactor Network (CRN) model, thus providing a relatively quick estimation of the combustion mechanisms under swirling combustion conditions. The simulation of NOx emissions indicates that Tian mechanism has better performance than the others compared. Hence, the Tian mechanism was selected as the most appropriate for further studies of ammonia/methane combustion through a set of experiments carried out at various equivalence ratios and pressure conditions. Finally, sensitivity and pathway analyses were also performed to identify important reactions and species under high pressurized conditions, areas that need more attention for model development and emission control in future studies.

Scalable Parameter Estimation for Genome-Scale Biochemical Reaction Networks.

Mechanistic mathematical modeling of biochemical reaction networks using ordinary differential equation (ODE) models has improved our understanding of small- and medium-scale biological processes. While the same should in principle hold for large- and genome-scale processes, the computational methods for the analysis of ODE models which describe hundreds or thousands of biochemical species and reactions are missing so far. While individual simulations are feasible, the inference of the model parameters from experimental data is computationally too intensive. In this manuscript, we evaluate adjoint sensitivity analysis for parameter estimation in large scale biochemical reaction networks. We present the approach for time-discrete measurement and compare it to state-of-the-art methods used in systems and computational biology. Our comparison reveals a significantly improved computational efficiency and a superior scalability of adjoint sensitivity analysis. The computational complexity is effectively independent of the number of parameters, enabling the analysis of large- and genome-scale models. Our study of a comprehensive kinetic model of ErbB signaling shows that parameter estimation using adjoint sensitivity analysis requires a fraction of the computation time of established methods. The proposed method will facilitate mechanistic modeling of genome-scale cellular processes, as required in the age of omics.

KERNEL: Enabler to build smart surrogates for online optimization and knowledge discovery

ABSTRACTKERNEL – A novel parameter-free surrogate building algorithm using Adaptive Neuro Fuzzy Inference System (ANFIS) is presented to provide an intelligent and robust technology to optimally estimate the configuration of ANFIS along with Sobol-based fast sample size determination (SSD) methodology. The proposed algorithm is capable of fine-tuning the existing knowledge base about the physics of the process in terms of human experience. It also enables knowledge discovery through a multi-objective optimization problem (MOOP) solved by non-dominated sorting genetic algorithm, NSGA-II, thus presenting machine-invented physics of the process. Experimentally validated polymerization reaction network model is considered and ANFIS surrogates are built using KERNEL. Surrogate-based optimization was found to be nine times faster than conventional optimization using the time expensive model, thus enabling its online implementation. Comparison of ANFIS with Kriging is also included.

ACRE: Absolute concentration robustness exploration in module-based combinatorial networks.

To engineer cells for industrial-scale application, a deep understanding of how to design molecular control mechanisms to tightly maintain functional stability under various fluctuations is crucial. Absolute concentration robustness (ACR) is a category of robustness in reaction network models in which the steady-state concentration of a molecular species is guaranteed to be invariant even with perturbations in the other molecular species in the network. Here, we introduce a software tool, absolute concentration robustness explorer (ACRE), which efficiently explores combinatorial biochemical networks for the ACR property. ACRE has a user-friendly interface, and it can facilitate efficient analysis of key structural features that guarantee the presence and the absence of the ACR property from combinatorial networks. Such analysis is expected to be useful in synthetic biology as it can increase our understanding of how to design molecular mechanisms to tightly control the concentration of molecular species. ACRE is freely available at https://github.com/ramzan1990/ACRE.

Learning reduced kinetic Monte Carlo models of complex chemistry from molecular dynamics.

We propose a novel statistical learning framework for automatically and efficiently building reduced kinetic Monte Carlo (KMC) models of large-scale elementary reaction networks from data generated by a single or few molecular dynamics simulations (MD). Existing approaches for identifying species and reactions from molecular dynamics typically use bond length and duration criteria, where bond duration is a fixed parameter motivated by an understanding of bond vibrational frequencies. In contrast, we show that for highly reactive systems, bond duration should be a model parameter that is chosen to maximize the predictive power of the resulting statistical model. We demonstrate our method on a high temperature, high pressure system of reacting liquid methane, and show that the learned KMC model is able to extrapolate more than an order of magnitude in time for key molecules. Additionally, our KMC model of elementary reactions enables us to isolate the most important set of reactions governing the behavior of key molecules found in the MD simulation. We develop a new data-driven algorithm to reduce the chemical reaction network which can be solved either as an integer program or efficiently using L1 regularization, and compare our results with simple count-based reduction. For our liquid methane system, we discover that rare reactions do not play a significant role in the system, and find that less than 7% of the approximately 2000 reactions observed from molecular dynamics are necessary to reproduce the molecular concentration over time of methane. The framework described in this work paves the way towards a genomic approach to studying complex chemical systems, where expensive MD simulation data can be reused to contribute to an increasingly large and accurate genome of elementary reactions and rates.

HSimulator: Hybrid Stochastic/Deterministic Simulation of Biochemical Reaction Networks

HSimulator is a multithread simulator for mass-action biochemical reaction systems placed in a well-mixed environment. HSimulator provides optimized implementation of a set of widespread state-of-the-art stochastic, deterministic, and hybrid simulation strategies including the first publicly available implementation of the Hybrid Rejection-based Stochastic Simulation Algorithm (HRSSA). HRSSA, the fastest hybrid algorithm to date, allows for an efficient simulation of the models while ensuring the exact simulation of a subset of the reaction network modeling slow reactions. Benchmarks show that HSimulator is often considerably faster than the other considered simulators. The software, running on Java v6.0 or higher, offers a simulation GUI for modeling and visually exploring biological processes and a Javadoc-documented Java library to support the development of custom applications. HSimulator is released under the COSBI Shared Source license agreement (COSBI-SSLA).

Steam pyrolysis of tetrabutyldiglycolamide: Part-II: Reaction network kinetic modelling

A Reaction Network model of 16 series-parallel reactions involving total 19 components is developed to represent the steam pyrolysis of tetrabutyldiglycolamide, a tridentate extractant used for selective extraction of actinides and lanthanides in nuclear industry, in a continuous tubular reactor. The kinetic parameters of the entire reaction network were optimized simultaneously and validated against the set of experimental results at different operating conditions. The reaction network model was successfully used for simulation of the pyrolytic reactor in General Process Modelling (gPROM) environment. The steam pyrolysis provides an efficient waste management method in nuclear as well as other industries treating concentrated organics waste streams that leaves no solid residue.

Entrainment of a Bacterial Synthetic Gene Oscillator through Proteolytic Queueing.

Internal chemical oscillators (chemical clocks) direct the behavior of numerous biological systems, and maintenance of a given period and phase among many such oscillators may be important for their proper function. However, both environmental variability and fundamental molecular noise can cause biochemical oscillators to lose coherence. One solution to maintaining coherence is entrainment, where an external signal provides a cue that resets the phase of the oscillators. In this work, we study the entrainment of gene networks by a queueing interaction established by competition between proteins for a common proteolytic pathway. Principles of queueing entrainment are investigated for an established synthetic oscillator in Escherichia coli. We first explore this theoretically using a standard chemical reaction network model and a map-based model, both of which suggest that queueing entrainment can be achieved through pulsatile production of an additional protein competing for a common degradation pathway with the oscillator proteins. We then use a combination of microfluidics and fluorescence microscopy to verify that pulse trains modulating the production rate of a fluorescent protein targeted to the same protease (ClpXP) as the synthetic oscillator can entrain the oscillator.

An efficient finite-difference strategy for sensitivity analysis of stochastic models of biochemical systems

Sensitivity analysis characterizes the dependence of a model's behaviour on system parameters. It is a critical tool in the formulation, characterization, and verification of models of biochemical reaction networks, for which confident estimates of parameter values are often lacking. In this paper, we propose a novel method for sensitivity analysis of discrete stochastic models of biochemical reaction systems whose dynamics occur over a range of timescales. This method combines finite-difference approximations and adaptive tau-leaping strategies to efficiently estimate parametric sensitivities for stiff stochastic biochemical kinetics models, with negligible loss in accuracy compared with previously published approaches. We analyze several models of interest to illustrate the advantages of our method.

Kinetic Model Development for Dehydration of 2,3-Butanediol to 1,3-Butadiene and Methyl Ethyl Ketone over an Amorphous Calcium Phosphate Catalyst

A reaction network and kinetic model for the dehydration of 2,3-butanediol (2,3-BDO) to 1,3-butadiene (1,3-BD) and methyl ethyl ketone (MEK) on an amorphous calcium phosphate catalyst are proposed. The kinetic parameters of the model were estimated using experimental data obtained from a laboratory-scale fixed-bed reactor operated under isothermal conditions. Experiments were performed using a mixture of 2,3-BDO, 3-buten-2-ol (3B2OL), and N2 as feed, at temperatures ranging from 304 to 334.5 °C and gas hourly space velocity (GHSV) ranging from 1780 to 2222 h–1. 2,3-BDO conversion varied from 6 to 100%; the selectivity of 1,3-BD ranged from 5 to 33 wt %, and the selectivity of MEK ranged from 31 to 34 wt %. Kinetic models based on the simple power law and Langmuir–Hinshelwood–Hougen–Watson model were developed to describe the dehydration of 2,3-BDO to 1,3-BD and MEK. Statistical and physicochemical criteria are used to contrast the performance of the two kinetic approaches. The power law model showed the h...

Emissions of a Wet Premixed Flame of Natural Gas and a Mixture With Hydrogen at High Pressure

It is generally accepted that combustion of hydrogen and natural gas mixtures will become more prevalent in the near future, to allow for a further penetration of renewables in the European power generation system. The current work aims at the demonstration of the advantages of steam dilution, when highly reactive combustible mixtures are used in a swirl-stabilized combustor. To this end, high-pressure experiments have been conducted with a generic swirl-stabilized combustor featuring axial air injection to increase flashback safety. The experiments have been conducted with two fuel mixtures, at various pressure levels up to 9 bar and at four levels of steam dilution up to 25% steam-to-air mass flow ratio. Natural gas has been used as a reference fuel, whereas a mixture of natural gas and hydrogen (10% hydrogen by mass) represented an upper limit of hydrogen concentration in a natural gas network with hydrogen enrichment. The results of the emissions measurements are presented along with a reactor network model. The latter is applied as a means to qualitatively understand the chemical processes responsible for the observed emissions and their trends with increasing pressure and steam injection.