Lengeler J W
Fachbereich Biologie/Chemie, Arbeitsgruppe Genetik, Universität Osnabrück, Germany.
Biol Chem. 2000 Sep-Oct;381(9-10):911-20. doi: 10.1515/BC.2000.112.
Complete genomes, far advanced proteomes, and even 'metabolomes' are available for at least a few organisms, e.g., Escherichia coli. Systematic functional analyses of such complete data sets will produce a wealth of information and promise an understanding of the dynamics of complex biological networks and perhaps even of entire living organisms. Such complete and holistic descriptions of biological systems, however, will increasingly require a quantitative analysis and the help of mathematical models for simulating whole systems. In particular, new procedures are required that allow a meaningful reduction of the information derived from complex systems that will consequently be used in the modeling process. In this review the biological elements of such a modeling procedure will be described. In a first step, complex living systems must be structured into well-defined and clearly delimited functional units, the elements of which have a common physiological goal, belong to a single genetic unit, and respond to the signals of a signal transduction system that senses changes in physiological states of the organism. These functional units occur at each level of complexity and more complex units originate by grouping several lower level elements into a single, more complex unit. To each complexity level corresponds a global regulator that is epistatic over lower level regulators. After its structuring into modules (functional units), a biological system is converted in a second step into mathematical submodels that by progressive combination can also be assembled into more aggregated model structures. Such a simplification of a cell (an organism) reduces its complexity to a level amenable to present modeling capacities. The universal biochemistry, however, promises a set of rules valid for modeling biological systems, from unicellular microorganisms and cells, to multicellular organisms and to populations.
至少对几种生物，如大肠杆菌而言，完整的基因组、先进得多的蛋白质组，甚至“代谢组”都已可得。对这些完整数据集进行系统的功能分析将产生大量信息，并有望让人理解复杂生物网络乃至整个生物体的动态过程。然而，对生物系统进行这样完整而全面的描述将越来越需要定量分析以及数学模型的帮助来模拟整个系统。特别是，需要新的程序，以便有意义地减少从复杂系统中获取的信息，这些信息随后将用于建模过程。在这篇综述中，将描述这种建模程序的生物学要素。第一步，必须将复杂的生命系统构建成定义明确且界限清晰的功能单元，其组成要素具有共同的生理目标，属于单个遗传单元，并对感知生物体生理状态变化的信号转导系统的信号作出反应。这些功能单元存在于每个复杂程度级别，更复杂的单元是通过将几个较低级别的要素组合成一个更复杂的单元而形成的。每个复杂程度级别都对应一个全局调节因子，它对较低级别的调节因子具有上位性。在将生物系统构建成模块（功能单元）之后，第二步将其转化为数学子模型，通过逐步组合，这些子模型也可以组装成更综合的模型结构。对细胞（生物体）的这种简化将其复杂性降低到适合当前建模能力的水平。然而，通用生物化学有望提供一套适用于对生物系统进行建模的规则，从单细胞微生物和细胞到多细胞生物体再到种群。
