Learning from the Steersman: A Natural History of Cybernetic Models

Abstract

Cybernetic models rely on optimal control heuristics to predict the effects of metabolic regulation on the dynamics of biochemical reaction networks. Over the past 30+ years, this fertile paradigm has brought forth scores of research publications and has witnessed diverse applications ranging from bioprocess optimization and control to the redesign of cellular hosts through rational metabolic engineering. This review traces the historical development of the cybernetic modeling framework, beginning with its philosophical presuppositions and then following its progress from first-generation "lumped" cybernetic models to biochemically detailed second-generation models to recent "hybrid" cybernetic models that combine aspects of both first-and second-generation models. The broad and lasting influence of the cybernetic modeling approach is due in large part to the fact that it has remained true to its foundational principles while adapting over time to meet the changing demands of biotechnology research. ■ INTRODUCTIONBiological systems are characterized by their ability to grow and reproduce under diverse environmental conditions. Isolates of the archaebacterial genus Pyrodictium, for instance, exhibit optimal growth at temperatures near 105°C in deep-sea hydrothermal vents, while the bacterium Thiobacillus ferrooxidans can tolerate acidities down to pH 1.0 in mine water runoff streams containing high levels of heavy-metal impurities. 1 Even organisms that exist in comparatively mild habitats must cope with frequent cycles of feast and famine along with a constantly changing menu of available nutrients. This astonishing resiliency is owed to a knack for regulating their metabolic machinery in ways that ultimately promote survival. Metabolism, the term used to denote the totality of chemical reactions in an organism, is enabled by a vast collection of enzymes with highly specific catalytic functions. Metabolic regulation is responsible for controlling the amounts and catalytic activities of these enzymes. The rates of metabolic reactions are thus subject to rigorous control aimed at protecting the interests of the organism.Accounting for this metabolic regulation in a rational manner is the main challenge faced by mathematical modelers who aim to predict the outcome of biological experiments based on a quantitative representation of the underlying cellular processes. The traditional kinetic modeling approach involves postulating biochemical and regulatory mechanisms that determine the rate of each enzymatic reaction, which can then be translated into corresponding kinetic expressions. Unfortunately, this requires a detailed knowledge of regulatory interactions and leads to a rapidly expanding set of kinetic parameters. The cybernetic modeling approach, on the other hand, views the complete biological system from an evolutionary perspective. On the basis of a general understanding of the metabolic objectives that an organism must satisfy in order to grow and survive, cybernetic control laws are postulated to reflect pu...

Dealing with indetermination in biochemical networks

“…The idea that a system, like a cell, seeks a biological objective is justified acknowledging the apparent goal-directedness of nature. Other computational approaches, like cybernetic modelling [21], have taken advantage of this concept, which has been formally defined with the word teleonomy. A teleonomic view of nature indeed considers that the genetic programs of all living organisms has been shaped by evolution in a way that highly performing phenotypes have been selected [21].…”

“…Other computational approaches, like cybernetic modelling [21], have taken advantage of this concept, which has been formally defined with the word teleonomy. A teleonomic view of nature indeed considers that the genetic programs of all living organisms has been shaped by evolution in a way that highly performing phenotypes have been selected [21]. Both FBA and cybernetic modelling have proved that this perspective can be used in different ways to simulate real biological behaviours.…”

“…It is worth highlighting that dFBA always considers the intracellular system at steady state, so it neglects all the dynamic fluctuations of internal metabolites. In the field of mechanism based models, the CM [21] closely follows the teleonomic perspective, producing a long history of progressive refinements of their technique. Besides teleonomy, the CM approach also assumes that (i) the topology of the network is completely known, that (ii)metabolic enzymes are regulated at the gene expression level or via allosteric mech-anisms, that (iii) some economic principles govern the way the cell implements these regulations, meaning that the cell, as if it were a rational investor, enhances those reactions that assure the highest yield in terms of the objective function.…”

New numerical simulation of the oscillatory phenomena occurring in the bioethanol production process

“…These models should be simplistic in the considered design. Through their formulization, There is no need to occupy huge information of the processes of biochemical but we need the capability to build the most uncomplicated potential map of reactions of metabolic [22]. Profoundly special enzymes are used to catalyse such reactions.…”

Regulation-Structured Dynamic Metabolic Model Provides a Potential Mechanism for Delayed Enzyme Response in Denitrification Process

“…As a key advantage, this approach facilitates the application of existing dynamic modeling platforms that have been developed to predict metabolic behaviors of single organisms (Song et al, 2014 ; Vasilakou et al, 2016 ). In this regard, the cybernetic approach developed by Ramkrishna and co-workers (Ramkrishna, 1983 ; Ramkrishna and Song, 2012 ; Young, 2015 ) provides an ideal platform for the enzyme-based microbial community modeling due to its rational description of metabolic regulation through the control of enzyme syntheses and activities.…”