School of Electrical Engineering and Computer Science The University of Newcastle, Callaghan, NSW, 2308, Australia.
Bioinformatics. 2010 Jan 15;26(2):283-4. doi: 10.1093/bioinformatics/btp652. Epub 2009 Dec 4.
This work was motivated by the need for an automated tool for discovery of genetic networks and the availability of extensive contextual protein-protein interaction information in the iHOP repository. At the moment, this information cannot be explored to its full potential due to the lack of software tools to reliably collect, process and display that information in a way that life scientists can quickly analyze genes of interest and search for potential interaction networks. Commercial tools can perform a similar job, but results appear to be less informative than those obtained using contextual information.
The Gene Interaction Miner (GIM) could successfully uncover complex network structures of protein-protein interactions for a test dataset composed of genes already related to Alzheimer's disease. That same set, when examined using two other analysis tools, namely STRING and Pathway Studio, resulted in incomplete protein-protein interaction networks, which indicate that the use of curated databases only gives a partial picture of the biological processes behind the disease.
The dataset used in this work and a running version of the software tool is available for download from the web site http://www.cs.newcastle.edu.au/~mendes/softwareGIM.html.
本工作的动机是需要一个自动化的工具来发现遗传网络，并且在 iHOP 存储库中有广泛的上下文蛋白质-蛋白质相互作用信息。目前，由于缺乏软件工具来可靠地收集、处理和显示这些信息，因此无法充分利用这些信息。生命科学家可以快速分析感兴趣的基因并搜索潜在的相互作用网络。商业工具可以执行类似的工作，但结果似乎不如使用上下文信息获得的结果有信息。
基因交互挖掘器（GIM）可以成功地揭示由已经与阿尔茨海默病相关的基因组成的测试数据集的复杂蛋白质-蛋白质相互作用网络结构。使用其他两个分析工具，即 STRING 和 Pathway Studio 对同一组进行检查，结果得到不完整的蛋白质-蛋白质相互作用网络，这表明仅使用已审核的数据库只能部分了解疾病背后的生物学过程。
本工作中使用的数据集和软件工具的运行版本可从网站 http://www.cs.newcastle.edu.au/~mendes/softwareGIM.html 下载。
