The BioGRID Database: An Open International Resource for Biological Interactions

复杂的生物体，如人类，是由数以万亿计的数百种不同类型的细胞组成的，这些细胞组成了组织和器官。每个细胞本身由基因组编码的数万种蛋白质组成；反过来，细胞的行为是由一个巨大的基因和蛋白质相互作用的网络决定的，这个网络控制着几乎所有的细胞功能，无论是分裂、基因表达、细胞形状和运动、发育，还是能量产生和代谢。无论是正常情况下，还是在疾病过程中出现问题时，理解细胞如何发挥作用的困难在于，细胞网络是如此错综复杂地相互联系在一起，以至于很难分析每个基因的作用。就像在生物化学中一样，当我们单独分析这些成分时，我们对每个单独的过程了解得很多，但不一定了解它们是如何协同工作的。相反，如果我们试图通过基因突变来去除这个谜题的关键部分，我们可能会失去整个过程的功能，从而无法辨别出问题基因的确切功能。不同基因的多重功能，以及细胞网络中可能出现的内置冗余，使这些困难进一步复杂化。系统生物学的方法是结合许多不同类型的测量，包括生物化学和遗传学，以了解每个单独的蛋白质和基因的累积效应。试图了解细胞和生物体如何运作的关键第一步是跟踪细胞中成千上万的相互作用。科学文献记录了全球数千名研究人员研究许多不同物种的结果。基因组序列分析以及对从酵母到人类等物种的特定基因功能的研究得出的深刻见解之一是，所有生物体实际上都是非常密切相关的。因此，在酵母细胞中运作的许多机制与在人类细胞中运作的机制基本相同。由于这个原因，对酵母、蠕虫、苍蝇和植物等“模式”生物的研究已被证明在理解人类疾病等方面提供了极其丰富的信息。事实上，对一种生物的基本发现必然会导致对所有物种的更好理解。理解的一个主要手段是阐明细胞的遗传和蛋白质相互作用。已经设计了各种技术来检测各种相互作用，并且许多这些方法现在可以以快速高通量的方式进行。尽管所有发现的相互作用都记录在科学文献中，但出现了一个问题，即没有这种相互作用的中央存储库，因此难以以系统的方式进行分析。为了解决这个问题，我们组建了一个由计算机程序员和生物学家组成的团队，他们负责整理数千种蛋白质和基因相互作用的科学文献，我们根据各种标准系统地对这些文献进行分类，然后永久地存放在一个名为BioGRID的开放访问在线数据库中（www.thebiogrid.org）。一个名为鱼鹰（Osprey）的图形界面使生物学家能够从这些相互作用中轻松地组装网络，从而推断出它们的功能。我们建议在爱丁堡大学建立BioGRID数据库的主要门户，并通过酵母、蠕虫、苍蝇、植物和人类细胞的基因和蛋白质相互作用的管理来扩大BioGRID/Osprey系统的容量。我们还将开发新的软件工具，以提高BioGRID/Osprey的性能，包括与爱丁堡系统生物学中心的其他复杂分析平台建立联系。BioGRID中的大型交互数据集将证明对国际学术生命科学界以及生物技术和制药工业具有不可估量的价值。

Complex organisms, such as humans, are composed of trillions of cells of many hundreds of different types, organized into tissues and organs. Each cell is itself composed of tens of thousands of proteins encoded by the genome; in turn, the cell's behavior is dictated by a vast network of gene and protein interactions that control virtually every cellular function, whether it be division, gene expression, cell shape and movement, development, or energy production and metabolism. The difficulty in understanding just how cells function, either normally, or when processes run awry in disease, is that cellular networks are so intricately interconnected that it is difficult to dissect the contribution of each individual gene. While analyzing the components individually, as in biochemistry, we learn much about each individual process, but not necessarily how they work together. Conversely, if we attempt to remove key components from the puzzle by genetic mutations, we may lose the function of the entire process, and thus not be able to discern the precise function of the gene in question. These difficulties are further confounded by the often multiple functions of different genes, and by the built in redundancies that can occur in cellular networks. The approach of systems biology is to combine many different types of measurements, both biochemical and genetic, in order to understand the cumulative effect of each individual protein and gene. A crucial first step in attempting to understand how cells and organisms function is to keep track of the many hundreds of thousands of interactions in the cell. The scientific literature records the results obtained by thousands of researchers around the globe, who study many different species. One of the profound insights to come from genome sequence analysis, as well as from the study of specific gene functions in species that range from yeast to humans, is that all organisms are in fact very closely related. Thus, many of the mechanisms that operate in a yeast cell are fundamentally the same as those in a human cell. For this reason, the study of 'model' organisms, such as yeast, worms, flies and plants, has proven extremely informative in understanding, for example, human disease. Indeed, fundamental discoveries in one organism invariably lead to a better understanding of all species. A primary means of understanding is to elucidate the genetic and protein interactions of the cell. A variety of techniques have been devised to detect various interactions, and many of these methods can now be carried out in a rapid high throughput fashion. Although all interactions discovered are recorded in the scientific literature, a problem has emerged in that there is no central repository for such interactions, which are thus difficult to analyze in a systematic manner. To solve this issue, we have assembled a team of computer programmers and biologists who curate the scientific literature for thousands of protein and genetic interactions, which we systematically classify by various criteria and then house permanently in an open access on-line database called the BioGRID (www.thebiogrid.org). An associated graphical interface called Osprey enables biologists to easily assemble networks from these interactions, and thereby deduce their function. We propose to establish a primary portal for the BioGRID database at the University of Edinburgh and to expand the capacity of the BioGRID/Osprey system through the curation of gene and protein interactions from yeasts, worms, flies, plants and human cells. We will also develop new software tools to enhance the performance of BioGRID/Osprey, including building links to other sophisticated analysis platforms at the Centre for Systems Biology in Edinburgh. The large interaction datasets housed in BioGRID will prove invaluable to the international academic life sciences community and to the biotechnology and pharmaceutical industries.