Miller L V, Krebs N F, Hambidge K M
Center for Human Nutrition, University of Colorado Health Sciences Center, Denver 80262, USA.
Adv Exp Med Biol. 1998;445:253-69. doi: 10.1007/978-1-4899-1959-5_16.
Compartmental modeling is a useful tool for investigating metabolic systems and processes. We and others have applied it to the study of zinc metabolism in humans. Because existing models could not be accurately fitted to our data, we have developed a new model of human zinc metabolism based on stable isotope tracer data from studies of five healthy adults. Multiple isotope tracers were administered orally and intravenously and the resulting enrichment measurement in plasma, erythrocytes, urine, and feces. These tracer kinetic data, along with other measured and calculated tracee and steady-state data, were used to develop the model. A single model structure composed of fourteen compartments was found to be suitable for all subjects. Model development and fitting of data and model for each subject were accomplished using the SAAM/CONSAM computer programs. The model development and fiting processes are described and exemplified using data from one of the subjects. While identifiability could not be demonstrated a priori due to the model's complexity, parameter statistics for the fitted models did show most parameters to be adequately identified a posteriori.
房室模型是研究代谢系统和过程的一种有用工具。我们和其他人已将其应用于人体锌代谢的研究。由于现有模型无法准确拟合我们的数据，我们基于对五名健康成年人的研究中获得的稳定同位素示踪数据，开发了一种新的人体锌代谢模型。口服和静脉注射多种同位素示踪剂，并测量血浆、红细胞、尿液和粪便中由此产生的富集情况。这些示踪剂动力学数据，连同其他测量和计算得到的被示踪物及稳态数据，被用于构建模型。发现由14个房室组成的单一模型结构适用于所有受试者。使用SAAM/CONSAM计算机程序完成了每个受试者的数据模型开发以及模型拟合。使用来自其中一名受试者的数据对模型开发和拟合过程进行了描述和举例说明。由于模型的复杂性，无法先验地证明可识别性，但拟合模型的参数统计确实表明大多数参数在事后得到了充分识别。
