Molecular Interaction Maps and Analysis of Bioregulatory Networks

为了更多地了解生物调控网络如何控制正常细胞和癌细胞的细胞周期，我们与一个跨学科的团队合作生成电子分子相互作用图(MIMS)，其中显示了正常生长期间和在扰乱细胞周期的条件下细胞周期调控途径的行为。这些努力有助于开发生物信息学工具来组织大量事实，包括对相互作用的调控分子、多蛋白质复合体、蛋白质修饰(例如磷酸化)等网络的描述。组织分子知识的主要障碍之一是缺乏一种共同语言，使科学家能够以清晰、标准化且最好是计算机可读的格式整合数据。为此，我们实现了分子相互作用图(MIM)语言，这是Kurt Kohn首先提出的一种图形化注释，它以图(分子相互作用图或MIM)的形式编码分子信息。这些MIMs被用来表示和分析分子相互作用，就像电路图被用来排除电子设备故障一样。研究人员通常用卡通式的图表描述生化途径，但这些对分子相互作用的表示通常是不完整和模棱两可的。例如，两个组分之间的箭头可能意味着数量的增加，活性的增加，或者一个分子对另一个分子的修饰。此外，生物调节网络中的酶通常是其他酶的底物，分子经常受到改变其结合或酶功能的修饰。此外，调节蛋白可以形成具有不同活性的多分子复合体，这取决于它们的组成和修饰。最后，调控分子中的每个结构域都可能有自己的结合、修饰和/或酶功能。因此，一个分子的活性和相互作用能力可能取决于它的修饰状态，以及它可能与之结合的其他分子。所有这些相互作用都必须考虑在内，才能充分理解系统。在MIM语言中，我们使用少量定义的、明确的图形符号来描述每种类型的分子相互作用。每个分子在图中的一个位置表示，分子之间的相互作用由连接线末端的箭头或条来指定。由于修饰后的分子和多分子复合体可能与原始分子具有不同的性质，因此每个相互作用的结果(如磷酸化分子或多分子复合体)被描绘为一个圆，或相互作用线上的节点。这些节点的处理方式允许它们形成更多交互并扩展网络。语言中使用的符号和约定，以及MIM的例子，可以在我们的网站上访问：http://discover.nci.nih.gov/mim，并在描述MIM语言原理的文章中访问。图形MIM语言允许同时查看涉及任何给定分子的许多相互作用。它可以描述生物调控网络中常见的相互竞争作用。感兴趣的研究人员可以从一个位置追踪给定分子的所有相互作用。读者可以在词汇表中查找分子，或者在电子(EMIM)图中，鼠标点击分子名称打开更多信息的链接。每个交互都标有一个指向带注释的描述的链接，其中包括指向引用文献的链接。感兴趣的研究人员可以阅读注释以获得关于每个分子相互作用的深入信息，或者浏览各种地图以熟悉细胞如何调节特定代谢过程的一般概念。例如，描绘DNA复制早期阶段的EMIM展示了参与该过程的分子之间所有可能的分子相互作用；其他地图代表了在细胞周期的特定阶段和对细胞压力的反应中发生的相互作用的子集。我们汇编和更新主要生物防治系统的地图，并努力将它们张贴在一个网站上，供公众使用，并以简明的方式将它们整合在一起。然后，我们可能会辨别出分子相互作用逻辑的常见模式，这些模式赋予了生物调控网络非凡的灵活性和健壮性。为了从MIM中描述的众多分子相互作用中阐明信号通路的逻辑，我们正在与一个多学科的研究小组互动，以开发基于MIM的计算机模拟。这些工具将阐明细胞控制DNA复制和细胞周期进程的过程，并可能帮助我们了解癌细胞中发生的细胞周期进程的扰动，并奠定这些细胞对抗肿瘤药物的敏感性。

To learn more about how bioregulatory networks control the cell cycle in normal and cancer cells, we collaborate with a cross-disciplinary team to generate electronic molecular interaction maps (MIMs), which show the behavior of cell cycle regulatory pathways during normal growth and under conditions that perturb the cell cycle. These efforts help develop bioinformatics tools that organize large collections of facts, including descriptions of networks of interacting regulatory molecules, multi-protein complexes, protein modifications (e.g., phosphorylations), etc. One of the main stumbling blocks to organizing molecular knowledge is the lack of a common language that allows scientists to integrate data in a clear, standardized, and preferably computer-readable format. To that end, we implemented the Molecular Interaction Map (MIM) language, a diagrammatic annotation first proposed by Kurt Kohn, which encodes molecular information in the form of diagrams (molecular interaction maps or MIMs). These MIMs are used to represent and analyze molecular interactions in the same way that circuit diagrams are used to trouble-shoot electronic devices. Investigators usually describe biochemical pathways in cartoon-like diagrams, but these representations of molecular interactions are often incomplete and ambiguous. For example, an arrow between two components could signify an increase in quantity, an increase in activity, or a modification of one molecule by the other. In addition, enzymes in bioregulatory networks are often substrates of other enzymes, and molecules are often subject to modifications that change their binding or enzymatic capabilities. Moreover, regulatory proteins can form multi-molecular complexes, which have different activities, depending on their composition and modifications. Finally, each domain within a regulatory molecule may have its own binding, modification, and/or enzymatic functions. Thus, a molecule's activity and interaction capabilities may depend on its modification state, and on the other molecules to which it may be bound. All of these interactions must be taken into account for a full understanding of the system. In the MIM language, we use a small number of defined, unambiguous graphical symbols to portray each type of molecular interaction. Each molecule is represented in a single place in a diagram, and interactions between molecules are specified by arrows or bars at the end of connecting lines. Because modified molecules and multi-molecular complexes may have different properties than the original molecules, the outcome of each interaction (such as a phosphorylated molecule, or a multi-molecular complex) is depicted as a circle, or "node" on an interaction line. These nodes are treated in a way that allows them to form more interactions and extend the network. The symbols and conventions used in the language, as well as examples of MIMs, can be accessed at our Web site: http://discover.nci.nih.gov/mim and in an article describing the principles of the MIM language. The graphical MIM language allows a simultaneous view of many interactions involving any given molecule. It can portray competing interactions, which are common in bioregulatory networks. An interested researcher can trace all the interactions of a given molecule from a single location. Readers can look up a molecule in a glossary, or in the electronic (eMIM) diagrams, a mouse-click on the molecule name opens links to more information. Each interaction is labeled with a link to an annotated description, which includes links to cited references. The interested researcher can read the annotations to gain in-depth information on each molecular interaction, or browse the various maps to become acquainted with the general concept of how cells regulate a particular metabolic process. For example, the eMIM depicting the early stages in DNA replication features all the possible molecular interactions between molecules involved in the process; additional maps represent subsets of interactions that occur during specific stages of the cell cycle and in response to cellular stress. We compile and update maps of the major biological control systems, and work to post them on a Web site for the use of the general public and to integrate them in a concise manner. We may then discern common patterns of molecular interaction logic that give bioregulatory networks their remarkable flexibility and robustness. To elucidate the logic of signaling pathways from the multitude of molecular interactions depicted in the MIMs, we are interacting with a multidisciplinary group of researchers to develop MIM-based computer simulations. Such tools will illustrate the processes by which cells govern DNA replication and cell cycle progression and may help us understand the perturbations in cell cycle progression that occur in cancer cells and underlie the sensitivity of these cells to anti-tumor drugs.