Model biosimulation

As was shown experimentally, micro-metabolites that are capable of acting as enzyme inhibitors may greatly affect product distribution. A biosimulation model can considerably assist drug approval by identifying physiological effects of drug metabolites too dilute in mammalian systems to be detected analytically. These results also show that the prediction of approximate product distribution requires implementation of a much more complex metabolic network, which is currently being done. 

To reveal this by conventional experiments or trials is almost impossible. However, a biosimulation model, or a set of models, can describe even very complex, interacting pathways and control systems quantitatively and give an estimate of the effect. Moreover, the models can indicate where and how a given metabolic disorder best can be corrected. 

The biosimulation models need not to be large and comprehensive. Large models are difficult to validate, because small changes in the model structure can lead to different outcomes. The models in the following sections are all of the type which could be called virtual experiments because the setup resembles an experimental setup. The advantage is that the biosimulation models easily can analyze hundreds of what if situations within a second or two. 

The present chapter has several aims. The first is to describe the glucose-insulin control system in a fashion that includes the most important physiological and biochemical processes governing the glucose metabolism. The second is to demonstrate how biosimulation models can be built and to show the quantitative arguments behind the setup of model equations. The third is to use these biosimulation models to reveal critical mechanisms in the glucose handling and to relate these mechanisms to experimental and clinical problems. 

The advantages of this approach are manifold. The breakdown into small models makes it easier to validate each individual facet, to overview its role in the total metabolism, and to relate it to more complex frameworks. It also becomes possible to combine different facets to model new scenarios and to predict the outcome of new experimental setups. The approach is also valuable in demonstrating how a biosimulation model can be created via a quantitative argumentation based on experimental results. 

This chapter has presented a series of small biosimulation models that describe different facets of glucose metabolism. The description is far from complete, and especially the role of the nervous system in this control is still poorly understood. 

Many results are presented as consequences of the biosimulation setups, and the experimental evidence for some of them may be questioned. This is not a weakness of the biosimulation. When a biosimulation model is constructed, it is based upon known experimental evidence, so the outcome of the model represents a series of conclusions based upon this evidence. The advantage is that the conclusions are quantitative and can point to new possible mechanisms that can be tested, to new experimental setups, and to new treatment targets. 

The biosimulation models in this chapter describe mainly healthy persons. In the case of diabetes some of the relations, regarding both carbohydrate and fat metabolism, are altered. One main change is a decreased insulin release for a given glucose increase. For the type 1 diabetes patient the decrease is fast, months to a few years, and leads rapidly to incompetent insulin production. For the type 2 diabetes patient the progress is much slower and takes decades. 

The development of new medicines is both lengthy and expensive. Many experiments and trials are necessary before the medicine reaches its final form. Integrating biosimulation models in the development can speed up the development process considerably and save much expense. This is instrumental for the development of individualized medicines, because otherwise the sales would be too small to cover the development costs. 

At present, few companies use biosimulation and only to a very limited extent There is therefore a large, unmet need for good biosimulation models and modelers in all branches of the pharmaceutical industry. But to satisfy the need is not... 

Predictive biosimulation is the use of computer modeling to put all the pieces of the biological puzzle together in a dynamic model that shows how they interact and work as a whole (see Chapters 6 and 22). It goes hand in hand with high-performance computing because it requires enormous computing resources. 

Fig. 3.1 Abstraction of substrate properties to common chemical principles is one of the basic concepts of the biosimulation approach (chemical abstraction). From the xenobiotic substrate molecule strategic positions for enzymatic attack are identified, since it is these that represent the reaction profile of a chemical compound in biological systems. The reaction profile determines regio-, chemo-, and stereoselectivity of enzymatic conversions. For these reasons model substrates representing the functionalities...

Fig. 3.3 Representative biotransformations serve as a basis for the development of a biosimulation-based approach. Model substrates for studying drug metabolism in the model eukaryote Saccharomyces cerevisiae ethyl acetoacetate (1), ethyl 4-chloro-acetoacetate (2) and ethyl 4,4,4-trifluoro-acetoacetate (3).

Biosimulation has the potential to drastically improve future drug development mathematical models may be used to obtain maximal information from experimental data, and can be used to pin-point potential drug targets, in a much more systematic manner than is the case today. 

Core-Box Modeling in the Biosimulation of Drug Action 5.2.4.2 Model Quality Analysis... 

The obtaining of different models for the same system that may easily be exchanged for each other is important not only in the biosimulations of a single system, for instance in the core-box modeling framework, but also when developing models for systems of systems. Such models are often referred to as hierarchical models, because they are formulated at different levels of complexity. For an hier-... 

It is clear, therefore, that core-box models with the additional characteristics of Eq. (11) (such as the insulin model in Section 5.3) have major potential in the future developments of hierarchical models describing larger systems. In this way, the potential drug targets predicted by the core-box model may be judged, not only by their quality tag, but also through their translated importance on the whole-body behavior. Both of these possibilities are very important to achieve the full potential of biosimulation of potential dmg targets. 

A real help for the industry and a possible breakthrough for biosimulation would be to find an easy way to validate each model and its results in a fashion that can compete with the high valuation of experimental and clinical results in the development process. 

Biosimulation has a dominant role to play in systems biology. In this chapter, we briefly outline two approaches to systems biology and the role that mathematical models has to play in them. Our focus is on kinetic models, and silicon cell models in particular. Silicon cell models are kinetic models that are firmly based on experiment. They allow for a test of our knowledge and identify gaps and the discovery of unanticipated behavior of molecular mechanisms. These models are very complicated to analyze because of the high level of molecular-mechanistic detail included in them. To facilitate their analysis and understanding of their behavior, model reduction is an important tool for the analysis of silicon cell models. We present balanced truncation as one method to perform model reduction and apply it to a silicon cell model of glycolysis in Saccharomyces cerevisiae. 

---