六．结束语

陈竺卫生部长上任前和上任不久就犯有四大错误：

1．错误地推进系统生物医学，它的原创成分太高，很多问题是未知数。是一个高投入，低收获的事。不符合中国国情。上海“系统生物医学研究中心”是个烂尾楼。所谓系统生物医学自己都弄不拎清，怎么能救中医。

2．鼓吹中医的整体观、辨证施治、治未病等核心理论思想。中医理论没有任何一点可取之处。所以日本才废医存药。

3．提出“亚健康”的这样极为错误的国家行动计划。

4．提出极为荒唐的预言：“建立融中西医学思想于一体的二十一世纪新医学。这种医学兼取两长，既高于现在的中医，也高于现在的西医。”陈竺把中国已经极为混乱的医疗派系思潮倒退了20年，回到1980年代的崔月犁时代。当时前中国卫生部长崔月犁说：“中医是我国能够为世界做出贡献的少数领域之一。”

中国卫生部长陈竺的这些荒谬的学术观点终究都将成为国际文明社会的笑柄。

（完）

附录1。崔芳，秦秋：“数字中医”铺设临床科研平台。健康报网2007年12月6日

附录2。王澄：原创性科学工作和模仿性科学工作的区别。新雨丝2007年12月3日

附录3。Denis Noble的文章的汉英对照译本

《科学》2002年三月Science 1 March 2002

第295卷，第5560号，第1678-1682页Vol. 295. no. 5560, pp. 1678 - 1682

综述Review

题目：心脏模型，从基因，细胞到整个心脏

Modeling the Heart--from Genes to Cells to the Whole Organ

作者：Denis Noble （英国）

地址：University Laboratory of Physiology, Parks Road, Oxford OX1 3PT, UK.

电邮：E-mail: denis.noble@physiol.ox.ac.uk

成功的生理学的分析需要能够了解细胞，器官，和系统这些关键部分彼此之间的功能性相互作用，以及这些相互作用在疾病时的变化。然而，我们想要知道的这些相互作用既不存在于基因组分里，也不存在于基因控制的个体蛋白中。那么它存在于何处呢？它存在于蛋白相互作用的水平。这个相互作用是指亚细胞，细胞，组织，器官，和系统结构之间的相互作用。因此，要确定健康的或疾病的这些逻辑关系，只有一条办法，就是记录下来每个部分的真实实验数据，再用电脑模拟化出它们之间的相互作用。今天，1。大量快速增加的生物学数据资料，2。已经建立的细胞模型，组织模型，和器官模型，3。大功率电脑硬件和演算法的出现，这三个方面的进步可以使我们有可能用定量的方法探索从基因水平到整体器官的生理功能与调节系统。本综述给大家展示了心脏模型的发展。21世纪的系统生理学成为高度量化的，也是电脑化的一个工作。

Successful physiological analysis requires an understanding of the functional interactions between the key components of cells, organs, and systems, as well as how these interactions change in disease states. This information resides neither in the genome nor even in the individual proteins that genes code for. It lies at the level of protein interactions within the context of subcellular, cellular, tissue, organ, and system structures. There is therefore no alternative to copying nature and computing these interactions to determine the logic of healthy and diseased states. The rapid growth in biological databases; models of cells, tissues, and organs; and the development of powerful computing hardware and algorithms have made it possible to explore functionality in a quantitative manner all the way from the level of genes to the physiological function of whole organs and regulatory systems. This review illustrates this development in the case of the heart. Systems physiology of the 21st century is set to become highly quantitative and, therefore, one of the most computer-intensive disciplines.

在过去的十年中，我们用新的技术获得的生物学数据之多已经完全超过了我们处理和分析它们的能力。基因组分已经提供给我们一本极厚的人体的“部件 （产品）目录”。蛋白组分试图给这些部件和详细结构作出定义。但是到目前为止，我们却没有一本“使用者手册”。如果有了这本手册，我们就能知道这些“部件”是怎样组合在一起，怎样彼此相互作用，才使生命得以维系或疾病得以发生。很多情况下，我们不知道基因和蛋白是怎样影响细胞，器官，以及系统的功能的，尽管从基因序列的相同处我们能够看到一些线索。进一步说，比如药物的治疗，如果我们只是知道药物仅仅在蛋白水平的作用机理是不够的。我们只有了解了在药物的作用下，蛋白和它的周围结构的作用，也就是和那些能产生更高层次的有关联的细胞机制，我们才能说我们正真懂了这个药物是如何作用的。如果没有这些整合起来的知识，我们甚至无法知道一个受体，一个酶，一个转移通道蛋白与哪种疾病状况有关。我们也会遇到药物副作用，而这些副作用仅仅从分子水平是无法预见的。（副作用是一个药物的多种作用的一部分。）

The amount of biological data generated over the past decade by new technologies has completely overwhelmed our ability to understand it. Genomics has provided us with a massive "parts catalog" for the human body; proteomics seeks to define these individual "parts" and the structures they form in detail. But there is as yet no "user's guide" describing how these parts are put together to allow those interactions that sustain life or cause disease. In many cases, the cellular, organ, and system functions of genes and proteins are unknown, although clues often come from similarity in the gene sequences. Moreover, even when we understand function at the protein level, successful intervention, for example, in drug therapy, depends on knowing how a protein behaves in context, as it interacts with the rest of the relevant cellular machinery to generate function at a higher level. Without this integrative knowledge, we may not even know in which disease states a receptor, enzyme, or transporter is relevant, and we will certainly encounter side effects that are unpredictable from molecular information alone.

如果只是检视基因组分数据资料，我们无法得到很多有关上述这些问题的解答。原因很简单，基因码只是为了蛋白的序列而设，所以这些基因码本身并不反映那些能产生器官功能的组成部分，也就是蛋白和其它细胞分子或亚器官相互之间的作用。这些基因码也不能告诉我们哪些蛋白在人的健康和疾病状态下对细胞和亚（小）器官功能起到关键作用。在生命机体中，很多这些相互作用的机理我们是不清楚的。只要有可能，大自然就会把这些我们想要知道的答案藏在分子的特殊化学结构里。在进化的过程中，使用这些化学结构的方式不仅是极为复杂的，而且极尽其用。如果拿人体的组织比作一部电脑，基因遗传密码的功能就像是电脑内事先编好的程序，这个程序运作起来之后能产生什么样的结果，基因密码并不知道。而蛋白组分才能够表明一些高一级的蛋白组合和蛋白相互作用的功能。Sydney Brenner说得好，“基因只能够决定它参与的那部分蛋白的特征，（基因的能力到此为止了）。而一个系统内那种被整合出来的特征则是由被复杂程序化的（蛋白）之间相互作用的结果。（后文比喻成“交响乐队”）。Sydney Brenner的意思是，因为生物学系统本身“复杂程序化”了这些（蛋白的）相互作用，所以，要了解他们，我们必须要“模拟程序化”他们，作为模拟分析的骨架。在这篇文章中，我要向大家介绍在心脏研究方面已经用模拟化来研究（蛋白）相互作用的进展。

Inspecting genome databases alone will not get us very far in addressing these problems. The reason is simple. Genes code for protein sequences. They do not explicitly code for the interactions between proteins and other cell molecules and organelles that generate function. Nor do they indicate which proteins are on the critical path for supporting cell and organelle function in health and disease. Much of the logic of the interactions in living systems is implicit. Wherever possible, nature leaves that to the chemical properties of the molecules themselves and to the exceedingly complex way in which these properties have been exploited during evolution. It is as though the function of the genetic code, viewed as a program, is to build the components of a computer, which then self-assembles to run programs about which the genetic code knows nothing, although proteomics can show us some aspects of the grouping and interaction of proteins (1). Sydney Brenner (2) expressed this very effectively when he wrote: "Genes can only specify the properties of the proteins they code for, and any integrative properties of the system must be `computed' by their interactions." Brenner meant not only that biological systems themselves "compute" these interactions but also that, in order to understand them, we need to compute them, and he concluded, "this provides a framework for analysis by simulation." In this review, I describe how far we have advanced in using simulation to understand these interactions in the case of the heart.

心脏的细胞（电脑）模型

心脏细胞模型已经作得非常精密了。因为有40年积累的数据和经验，这些经验来自实验工作和电脑模拟工作两点之间反复地（确认和建立）。现在我们已经有了所有的主要类型的心肌细胞模型。在特殊基因的表达方面，我们（的模型）现在也已经能够表达各种变量，比如，跨心室肌，窦房结的中心和周围之间，和心房内（这些特殊的心肌细胞）。这些变量对于我们了解整个心脏的表现，比如心电图的表现，和分析心律产生的方式，是极为重要的。心脏细胞模型能表达的变量对于疾病的研究也很重要，比如，心衰，能够让我们看到改变基因表达后心衰的特征。

Cellular Models of the Heart

Models of heart cells have become highly sophisticated and have benefited from four decades of iterative interaction between experimental and simulation work (3). Models of all the main types of cardiac myocyte exist (in many cases there are multiple models of the same cell type), and we are now able to represent the variations in the expression of particular genes, for example, across the ventricular wall (4), between the center and periphery of the sinoatrial node (5), and within the atrium (6). These variations are of fundamental importance in understanding global phenomena such as the electrocardiogram (ECG, see Fig. 1), and for analyzing the way in which cardiac rhythm is generated. They are also fundamental to understanding disease states, some of which, like heart failure (7), can be characterized by alterations in gene expression profiles.

--------------------------------------------------------------------------------

图片说明（略） Fig. 1. Reconstruction of cardiac ventricular transmural action potential shapes attributable to variations in gene expression levels and the insertion of these cellular models into a 3D model of the ventricular wall capable of reproducing the T wave of the ECG. Left, supercomputer reconstruction of electrical field (color coded) when ventricular wall wedge is inserted into a conducting medium. Middle, in silico models of endocardial, mid-myocardial (M cell) and epicardial cells together with the reconstructed ECG obtained from the wedge model. Right, Experimental recordings of dog ventricle (34, 35). The in silico records (left and middle) are from the CardioPrism cardiac safety assessment program of Physiome Sciences, Inc. (36).

--------------------------------------------------------------------------------

和基因水平的联系

非常重要的是，能够模拟蛋白功能特点的那些模型能够让我们触及到基因水平。比如，我们可以（记录）重建特殊突变基因的效果。这些突变带有为蛋白功能改变的特点。Markov模型就是这样的一个例子。Markov模型是心肌钠离子通道模型，用这种模型可以建立不寻常的变异的钠离子通道。这种模拟的变异叫做KPQ变异，去除了三个氨基酸，lysine-proline-glutamine赖氨酸脯氨酸谷氨酸，结果影响了通道的失活功能。这种现象在先天性长QT综合征中出现。这个模型能演示变异后钠通道可以“起死回生”，重新开放，甚至非常活耀。这是因为我们暂时中止了通道失活，从而产生了动作电位平台期的持续钠内流。接着引起复极化时间的延长，最后产生了一个早期除极化后的心动过缓现象。这个现象和长QT综合征病人在睡眠时和松弛时发生心动过缓的临床表现是一致的。

Linking to Genetics

An important strength of models based on reconstructing the functional properties of proteins is that it is possible for the models to reach down to the genetic level, for example, by reconstructing the effects of particular mutations when these are characterized by changes in protein function. An example of this approach is the use of multistate (Markov) models of the cardiac sodium channel (8) in which models of the wild-type and of a mutant sodium channel were formulated and validated. The simulated mutation was the KPQ mutation, a three-amino acid (lysine-proline-glutamine) deletion that affects the channel inactivation and is associated with a congenital form of the long-QT syndrome, LQT3. The simulations showed that mutant channel reopenings from the inactivated state and channel bursting due to a transient failure of inactivation generate a persistent inward sodium current during the action potential plateau in the mutant cell. This causes major prolongation of repolarization and the development of arrhythmogenic early afterdepolarizations at slow pacing rates, a behavior that is consistent with the clinical presentation of bradycardia-related arrhythmogenic episodes during sleep or relaxation in LQT3 patients.

另外一种钠通道变异模型目前多少属于一种“错觉”变异模型。这种变异可以影响到电压依赖的钠通道失活，这是原发性心室纤颤的原因之一。它的原理是，稍稍改变一点电压依赖的通道失活（能力），就可以在早期去极化后产生致死性的心律失常（心室纤颤）。早期去极化后也可引起充血性心衰的心律失常。这个过程可以被电脑模拟，模拟的数据来自于实验室里记录的基因表达的变化，这涉及到好几个转递蛋白。

Another sodium-channel mutation that has been, at least partially, reconstructed is a missense mutation that affects the voltage dependence of sodium-channel inactivation; it is responsible for one form of idiopathic ventricular fibrillation [the Brugada syndrome (9)]. In this case, small shifts of the voltage dependence of inactivation generate early afterdepolarizations that may underlie fatal arrhythmia (10). Early afterdepolarizations are also responsible for the arrhythmias of congestive heart failure. This process can be modeled on the basis of experimentally determined changes in gene expression for several of the transporter proteins involved (7).

上述举的例子告诉我们细胞模型的作用。它能够演示基因和离子通道异常后产生心律失常的过程。这种通道异常可以是行为的，也可以是表达的。就目前我们对基因学的了解，今后十年这些使用模型的研究依然重要。能把（小小的）基因组分的表现联系到生理效果，这是电脑化的生物学令人兴奋的工作之一。

These examples highlight the ability of cellular models to reconstruct the arrhythmogenic consequences of genetic and ion-channel abnormalities either of behavior or of expression levels. Given the present explosion of genetic information, such studies will continue to be at the forefront of modeling efforts in the next decade. Connecting the genome to physiology is one of the exciting prospects for computational biology.

与直觉相反

复杂模型系统有个特点，就是常常是与直觉相反。因为当一件事超过一定的复杂程度时，光凭想像（而不是从第一手直接获得的经验）不仅是不适当的，甚至可以误导。有一个好的例子可以说明这个观点，这个例子就是给细胞模型加上心肌缺血的生化变化的内容。结果这样做就可以模拟心律失常。在细胞内的钠和钙离子浓度都很高的条件下，如果我们推迟了去极化后期，细胞内的钙离子就活跃起来。

钙离子的活跃接着作用在钠钙交换泵上，产生了内向电流，细胞提前兴奋。如果在代谢损伤如心肌缺血的病理状况下，对钠钙泵的向上调节和向下调节，这个现象就具有与直觉相反的特征。这个转递通道蛋白现在是抗心律失常药物治疗的研究重点。电脑模型在分类和评估这些抗心律失常药物的作用机理上起到很大作用。因为它解答了那些转递蛋白变化顺序的复杂现象。

另外一个做得比较多的是心肌的机械和电的反馈模型。它产生了与直觉相反的实验结果：心肌的收缩能够影响它的电活动。这个现象已经被精确的实验和电脑化的工作所证实。有些实验结果无法预期，他们与细胞数量变化后的作用有关。而细胞数量变化又与很多疾病有关。这类研究就是下一步要开展的。回过头来看，如果我们不是反复工作在实验和电脑模型之间，我们就不可能揭示那样复杂的生理学内容。

Counterintuitive Predictions

Characteristically, the results of modeling complex systems are frequently counterintuitive. This occurs because, beyond a certain degree of complexity, armchair (qualitative) thinking is not only inadequate for understanding such systems, it can even be misleading. A good example of this comes from the extension of cellular models to include some of the biochemical changes that occur during ischemia (11). This work succeeds in reconstructing arrhythmias attributable to delayed afterdepolarizations that arise as a consequence of intracellular calcium oscillations in conditions in which intracellular concentrations of sodium and calcium become excessive. These oscillations generate an inward current carried by the sodium-calcium exchanger that can lead to premature excitation of the cell. This work has led to counterintuitive predictions concerning up- and down-regulation of sodium-calcium exchange in disease states involving metabolic damage, such as cardiac ischemia (12). This transporter is currently a focus of antiarrhythmia drug therapy. Simulation is playing an important role in clarifying and assessing the mechanism of action of such drugs, by unraveling the complex changes that occur as a consequence of the change in transporter activity.

Another area in which modeling has been rich in counterintuitive results is that of mechano-electric feedback, in which the contraction of the heart influences its electrical properties. This feedback mechanism has been unraveled in elegant experimental and computational work (13). Some of the results, particularly on the actions of changes in cell volume (which are important in many disease states) are unexpected and have been responsible for determining the next stage in experimental work. Indeed, it is hard to see how such unraveling of complex physiological processes can occur without the iterative interaction between experiment and simulation.

评估和预测药物的作用

药物是和蛋白起作用的。 蛋白质可以作为受体，通道，转递通道，和酶。能够影响蛋白的结构和功能的模型对于评估和预见药物的作用是很有用的。 美国FDA已经用模型来评估药效了。我们期待随着电脑模型的内容越来越复杂，电脑的功能越来越大，我们期待生物学模型的使用也就越来越广泛。药物对心脏的安全的评估就是在使用这些模型作为手段。自1998年以来，美国从市场上撤回的药物中的一半都是因为发现了那些药物对心脏不安全。先是心电图有变化，而后发生心律失常。（撤药）是巨大的浪费。实际上 ，所有的心肌复极化的离子转递通道的模型都已经作出来了。当把这些模型放入一个三维的心脏组织模型时，甚至非常逼真的心电图的T波模型也能得到。因此，我们就可以用silico模型来筛选药物。我们不得不用电脑模型的原因之一是心电图对于潜在的心律失常的发现很不可靠。不同的分子和细胞的良性或恶性的作用都可以引起QT段和T波的相同的变化。所以必须要了解从每一个离子通道的变化怎样能影响到心电图的过程。 随着我们在把精确的细胞模型整合到具有解剖细节的心脏器官模型上获得了更多的经验，这个目的就可以实现。

Assessing and predicting drug actions.

Drugs act on proteins such as receptors, channels, transporters, and enzymes. Models that simulate effects of perturbing protein structure and function are therefore highly relevant to assessing and predicting drug actions. Simulations have already been used in assessing drug action by the U.S. Food and Drug Administration, and we can expect use of such biological models to increase greatly as their complexity and power grows (14, 15). One obvious use in the case of the heart is in assessing the cardiac safety of drugs. It should be noted that half the drug withdrawals that have occurred since 1998 in the USA when drugs have come on the market have been attributable to cardiac side effects, often in the form of effects on the ECG and consequent arrhythmias. This is a large and very expensive form of attrition. Because virtually all the ion transporters involved in cardiac repolarization are now modeled and because very realistic simulations of the T wave of the ECG can be obtained when these models are incorporated into three-dimensional (3D) cardiac tissue models, it is clearly becoming possible to use in silico screens for drug development. One of the reasons that this is necessary is that the ECG is, unfortunately, an unreliable indicator of potential arrhythmogenicity. Similar changes in form of the QT interval and T waveform can be induced by very different molecular and cellular effects, some benign, others dangerous. We need to understand and predict the mechanisms all the way from individual channel properties through to the ECG. This goal is within reach, particularly as we acquire more experience of the incorporation of accurate cellular models into anatomically detailed organ models (see below).

模型的另一个作用是筛选药物的多种作用。几乎没有什么心脏药物只结合一个受体。大多数药物常常影响两个，三个或更多的受体和离子通道。那些作用于钠泵交换通道的药物都是影响多个受体的。需要指出的是药物的多个作用点很可能（对病人）有好处。我们也期待多受体的药物作用（比单一受体的药物）更好。我也期待通过这些能发现多受体的抗心律失常药物。人类的心功能的调节是很多种作用在一起的过程，特别是由G蛋白耦合受体的调节。在寻找人类疾病的自然调节的秘密的时候，我们必须更加精细地了解蛋白们的“乐队”的（有程序的）表演。我们先要模拟它，才能了解到它的复杂性，才能懂得那些多种作用发挥出来的功能。

Another use of simulation in drug discovery is screening drugs for multiple actions. Very few drugs that act on the heart bind to just one receptor. It is much more common for two, three, or even more receptors or channels to be affected. This is particularly true for drugs that act on the sodium-calcium exchanger (16). An important point to realize here is that multisite action may actually be beneficial. Many multireceptor drug actions are expected to be beneficial. I predict that this will be one of the ways in which more rational discovery of antiarrhythmic drugs may occur. In regulating cardiac function, nature has developed many multiple-action processes, particularly those regulated by G protein-coupled receptors. In seeking more "natural" ways of intervening in disease states, we should also be seeking to play the orchestra of proteins in more subtle ways. We need simulation to guide us through the complexity and to understand multiple action functionality.

把细胞模型整合到整体器官模型中

对于怎样发展生物学模型，已经有相当多的争论。 有人说要“从下向上（从基因到整体器官）”，有人说要“从上向下”，也有人说把两个方法部分结合起来。比较一致的意见是“从中间向上下发展”。这个建议基于中间部分的（已经完成的）生物学数据资料特别多，用这些数据做成模型后再向“上”或“下”伸延。说到心脏的研究，有两个部分的数据很多，一个是细胞水平的研究，另一个是三维的整体心脏的研究。如果把这两个部分的模型互相延伸，一定很有意义。心室的解剖学细节的模型，显示出心肌纤维的走行方向和肌层的结构。这些已经被结合在细胞模型中，来模拟心脏的电和机械力的作用。

Incorporation of cellular models into whole-organ models.

There has been considerable debate over the best strategy for biological simulation, whether it should be "bottom-up," "top-down" or some combination of the two [see discussions in (17, 18)]. The consensus is that it should be "middle-out," meaning that we start modeling at the level(s) at which there are rich biological data and then reach up and down to other levels. In the case of the heart, we have benefited from the fact that, in addition to the data-rich cellular level, there has also been data-rich modeling of the 3D geometry of the whole organ (19, 20). Connecting these two levels has been an exciting venture (21, 22). Anatomically detailed models of the ventricles, including fiber orientations and sheet structure, have been used to incorporate the cellular models in an attempt to reconstruct the electrical and mechanical behavior of the whole organ.

图2是激活波的传播图。这些图被心脏超微结构影响，电的传导最容易沿着心肌纤维－肌层轴的走向，这个电脑模拟结果和从狗的心脏多电极记录结果十分相同。图1是我在前面提到的模拟心电图的末期相，使用的是详细的复极化扩散的数据。准确地模拟去极化波可以模拟心电图的心室兴奋的早期相，也就是QRS波。这样一来，就把窦房结，心房，传导系统整合到整个心脏。这样，我们正在期待着第一个完整心脏生理过程的成功模拟，完整是指能从蛋白功能（的变化）一下子（解释到）到常规临床现象。现在已经做到，把整个心室模型结合到了电脑的虚拟的人体，包括不同组织的电传导功能。这样做就可以延伸到涉及到多导胸前和肢体记录的身体外部模拟。

Still pictures from a simulation in which the spread of the activation wavefront is reconstructed are shown in Fig. 2. This is heavily influenced by cardiac ultrastructure, with preferential conduction along the fiber-sheet axes, and the result corresponds well with that obtained from multielectrode recording from dog hearts in situ. I referred earlier (Fig. 1) to work that reconstructs the later phases of the ECG using detailed reconstruction of the dispersion of repolarization. Accurate reconstruction of the depolarization wavefront promises to provide reconstruction of the ECG during the early phases of ventricular excitation, i.e., the QRS complex, and as the sinus node, atrium, and conducting system are incorporated into this whole heart, we can look forward to the first example of reconstruction of a complete physiological process from the level of protein function right up to routine clinical observation. The whole ventricular model has already been incorporated into a virtual torso (23), including the electrical conducting properties of the different tissues, to extend the external field computations to reconstruction of multiple-lead chest and limb recording.

--------------------------------------------------------------------------------

图片说明（略） Fig. 2. Spread of the electrical activation wavefront in an anatomically detailed cardiac model (21). Earliest activation occurs at the left ventricular endocardial surface near the apex (left). Activation then spreads in endocardial-to-epicardial direction (outward) and from apex towards the base of the heart (upward, middle frames). The activation sequence is strongly influenced by the fibrous-sheet architecture of the myocardium, as illustrated by the nonuniform transmission of excitation. Red, activation wavefront; blue, endocardial surface.

--------------------------------------------------------------------------------

血流和冠状动脉循环

McQueen和Peskin建立了精准的心脏血流模型，包括心脏的腔和瓣膜的动作。这个模型已经用在舒张期的机械功能研究。冠状动脉循环中的血流模式也已建立。心肌缺血是严重心衰和死亡的主因。它也是很多多发病因素的疾病的好例子。几乎很少有疾病能够被归因于单个基因或蛋白的功能障碍。如前文所说，在模型上，当（缺血）心肌细胞能量供应减少后，细胞的代谢和电生理改变过程的模拟已经达到了可以复制某些心律失常的高水平。造成心肌能量供应减少的第一步是阻断冠脉。还有一个例子，不同的具有丰富数据资料的部分建起的模型，这个模型能产生让人非常兴奋的功能整合信息。图3是一些有名的冠脉循环的模型。这些研究冠脉血流的模型是从跳动的心室记录的。因此，模型也包括了心室的变形，这是机械对血流的影响。

Blood flow and the coronary circulation.

Blood flow within the chambers of the heart, including the movement of valves, has been elegantly modeled by McQueen and Peskin (24) and this has been extended to the study of diastolic mechanical function (25). Blood flow within the coronary circulation has also been modeled (26).

Ischemic heart disease is a major cause of serious incapacity and mortality. It is also a good example of the multifactorial character of most disease states. Very few diseases are attributable to a single gene or protein malfunction. As noted above, cellular reconstructions of the metabolic and electrophysiological processes that occur following deprivation of the energy supply to cardiac cells have already advanced to the point at which some arrhythmic mechanisms can be reproduced. The initiating process in such energy deprivation is restriction or block of coronary arteries. This is another example where modeling at different data-rich levels is holding out the prospect of very exciting integration of function. Some of the spectacular modeling of the coronary circulation are shown in Fig. 3 (26). These are stills from a simulation in which the blood flow through an anatomically detailed model of the coronary circulation is computed while the ventricles are beating. The simulation, therefore, also included the deformation that occurs as mechanical events influence blood flow.

--------------------------------------------------------------------------------

图片说明（略） Fig. 3. Flow calculations coupled to the deforming myocardium. The color coding represents transmural pressure acting on the coronary vessels from the myocardial stress (dark blue, zero pressure, red, peak pressure). The deformation states are (from left to right) zero pressure, end-diastole, early systole, and late systole (26).

--------------------------------------------------------------------------------

这个模型已经被用来研究把主要冠状动脉阻断后的血流变化。这项工作是把这个模型整合到细胞和组织水平的缺血模型上。如果我们能用心律失常的细胞水平的机理来解释“为什么本来是正常有序的心跳突然变成了致死性的心室颤动”，那么，整合生理学电脑模拟的新时代就来临了：人类能够全方位地模拟心绞痛了。

This model has already been used to investigate the changes in blood flow that occur following constriction or blockage of one of the main arterial branches, and work is in progress to connect this to the modeling of ischemia at the cell and tissue level (see Fig. 4). If we can also connect the cellular mechanisms of arrhythmia to the processes by which regular excitation breaks down into the multiple wavelets of ventricular fibrillation (27) then yet another "grand challenge" for integrative physiological computation will come within range: the fullscale reconstruction of a coronary heart attack.

--------------------------------------------------------------------------------

图片说明（略） Fig. 4. Left, the coronary circulation model shown in Fig. 3 has been subjected to a constriction of one of the main branches leading to blocked blood flow in the regions colored blue. Right, simulation of ectopic beats in a Purkinje fiber model in conditions of calcium overload of the kind that occurs in ischemic tissue. Oscillatory calcium changes (bottom) induce inward sodium-calcium exchange current (middle) leading to initiation of action potentials (above). Linking these two levels of modeling to create a complete model of coronary heart attack is one of the grand challenges requiring massive computer power. [Top panel kindly provided by N. Smith. Bottom panel specially prepared for this review using the DiFrancesco-Noble 1985 Purkinje fiber model (37) as follows. To simulate sodium/calcium overload, [Na]i was increased from 8 to 12 mM. The first action potential is evoked by a current pulse. The second two are initiated by calcium oscillations. Note that the rise in [Ca]i and the flow of inward Na-Ca exchange current occur before the depolarization.]

--------------------------------------------------------------------------------

我用“巨大的挑战”这个词是很慎重的。因为这样一个工作首先需要大型电脑。一次完整心脏的模拟实验需要超级电脑工作很多小时。进一步的工作如果没有电脑的超大容积将很难完成。这个工作可能会打破Moore定律，Moore定律是指电脑的能力每18个月翻一倍。IBM的蓝基因计划就会是这样。

The term "grand challenge" is chosen deliberately. This kind of work requires massive computer power. The whole organ simulations described here require many hours of computation using supercomputers. (By contrast, the single-cell models can be run faster than in real time on a PC or laptop!) Future progress will be determined partly by the availability of computing capacity. It is significant therefore that attempts to break Moore's law (computing power doubles every 18 months) are in progress, notably that of IBM's blue gene project (28).

未来：从基因组分，到蛋白组分，到生理学组分

电脑模型生物学系统是一个重要技术。它可以组织和整合极大量的生物学信息。尽管这篇文章只涉及到心脏的模型，可是生物学模拟模型现在已经被应用到广泛的领域，这包括各种通路，细胞和系统。在我们要利用快速基因排序和蛋白定谱的数据来创造生理学组分的过程中，silico（电脑）生物学模型在医学和制药的领域中的应用会更越发突出。

对模型的使用和建立是从每天的实验室工作中获得的。电脑模拟和实验室工作必须并行。这个工作的参与研究人员越多，进步将越快。也是由于在研究的具体内容的针对性不同，模型过于复杂，所以几乎不能提供给其它实验室用。我们希望今后能够发展出通用的模型，让不懂模型建立的人也能使用它。改善现有模型。现在也在网上可以看到越来越多的模型的介绍。并且，模型的交流和语言正在发展中。这些事都能做到的话，我们就信心十足地期待整合的细胞模型，器官模型，和系统模型的巨大的发展。用不了多少年，我们就离不开这些生物研究的模型了。药物研究更是如此。当这些电脑模型联网和功率增大以后，药物研究会有很大的变化。就生物学的内容来说，这种变化会是非线性的。（比线性要快。）因为被模型化的因素的相互作用的增加比研究因素的个数增加快。21世纪的生物学被期待为巨大量化的一门学科。电脑化将成为生物学主要内容。

The Future: From Genome to Proteome to Physiome

The computer modeling of biological systems is an important technique for organizing and integrating vast amounts of biological information. Although this review has focused on modeling of the heart, it is important to note that biological simulation is now being done for a wide range of pathways, cells, and systems (29). The role of in silico biology in medical and pharmaceutical research is likely to become increasingly prominent as we seek to exploit the data generated through rapid gene sequencing and proteomic mapping (1) through to creating the physiome (30, 31).

However, progress will be significantly enhanced by enabling ever greater numbers of researchers to use and verify models in the course of their everyday experimental work [for simulation and experiment must go together (3)]. It has been extremely difficult to transfer models between research centers, or to extend existing models so that more complex models can be constructed in an object-oriented or modular fashion. This process will be enhanced by the development of uniform standards for representing and communicating the content of models, and by the wide distribution of software tools that permit even nonmodelers to access, execute, and improve existing models. Increasingly, publication of models is accompanied by their availability on Web sites. Also, the process of establishing standards of communication and languages is developing (32, 33).

Once this is achieved, we can confidently predict an explosion in the development of integrated model cells, organs, and systems. In a few years' time we shall all wonder how we ever managed to do without them in biological research. For drug development, there will certainly be a major change as these tools come on line and rapidly increase in their power. This will grow in a nonlinear way with the degree of biological detail that is incorporated. The number of interactions modeled increases much faster than the number of components. Biology is set to become highly quantitative in the 21st century. It will become a computer-intensive discipline.

上一页  [1] [2] [3] [4] [5] [6] [7]  下一页