Merging genomic control networks and soil-plant-atmosphere-continuum models
Citation: Welch, S.M., Roe, J.L., Das, S., Dong, Z., He, R., Kirkham, M.B. 2005. Merging genomic control networks and soil-plant-atmosphere-continuum models. Agricultural Systems, 86(3):243-274.[PDF]
Merging genomic control networks and soil-plant-atmosphere-continuum models
S.M. Welch, J.L. Roe, S. Das, Z. Dong, R. He, M.B. Kirkham
ABSTRACT
Advances in genomic science make it desirable to include genomic controls in soil-plant-atmosphere-continuum (SPAC) models by methods proposed in this paper. Molecular genetic concepts suggest that a differential equation similar to ones used in neural networks can be used to model single-gene elements of larger systems. Natural modifications to the equation incorporate temperature dependency. Multi-gene components based on this element function as Boolean logic gates, linear arithmetic units, delays, differentiators, integrators, oscillators, coincidence detectors, and bi-stable devices. Related genetic circuitry from real organisms is shown. Genomic integration with SPAC models entails whole-plant modeling with realistic morphology. Plants are networks of parts, iterated in time and space under genetic control, that induce and modulate conservative SPAC mass/energy flows. Network developmental rules can be stated as Lindenmayer grammars whose symbols represent plant parts programmed as software objects. A structure is presented for simulators based on these concepts. The discussion argues that prior object-oriented plant modeling approaches (i) do not reflect how plants actually develop morphologically and (ii) may represent processes in tactically unwise ways at a time when genomics is advancing knowledge of process interactions. Finally, genomics and expanding computing power redefine concepts of model “simplicity” and “complexity” to favor increased realism.
文章摘要
基因组科学的进步使得将基因组控制加到土壤-植物-大气连续体（SPAC）模型成为可能。本文将提供了相应的方法。分子遗传学的概念表明，在神经网络中使用的微分方程可以用来模拟大型复杂系统中的单基因元素。在方程中纳入温度的依赖性是一种自然延伸。基于这种但基因元素的多基因此组件可以用作布尔逻辑开关、线性算术单元、延迟、微分子、积分子、振荡子、巧合探测器、和双态稳定装置。本文同时展示了真实生物体的关联遗传电路。整合了基因组信息的SPAC模型将使模拟真实的全株形态成为可能。植物是有部件组成的网络，在遗传控制下随着时间和空间不断迭代，从而诱发和调节保守的SPAC质量和能量流。Lindenmayer语法可以用来表达网络展开的规则，其中植物部件可用符号来代表，在编程时当作对象。基于这些概念，本文提供了一个可以作为模拟器的构架。随着基因组学进一步促进了我们对生理过程之间的相互交互的理解，我们认为先前的面向对象的植物建模方法有如下不足（1）不反映植物的实际形态发育和（2）表达生理过程的方法不够现实。最后，基因组学和计算能力的不断扩展将重新定义模型的“简单性”和“复杂性”，有利于增加模型的真实性。
