Rodeo: A Platform for Discovery and Analysis of Protein Network Motifs

哈佛大学获得了一笔赠款，用于建立一种新的公开可用的信息学工具，名为Rodeo，它将允许整合与破译蛋白质-蛋白质相互作用网络相关的数据，并将提供许多分析此类数据的方法，包括图论数学。Rodeo将允许集成和分析几个数据库，包括来自50多个基因组的一组全面的同源物，来自人类的一组新的蛋白质-蛋白质(p-p)相互作用数据，以及来自其他生物的超过70,000个p-p相互作用，来自NCBI主持的新实验和实验的大量基因表达数据集，以及来自一种有机体(橡子蠕虫)的近10万个表达序列标签的数据库，这可能是理解现代脊椎动物进化出的发育复杂性的关键。竞技场将不仅允许询问一个有机体的数据，而且允许询问多个有机体的数据，提供了一种搜索一致进化模式的方法，这是理解过程的关键一步。竞技表演还将允许可视化的主题，以字面上‘看到’网络复杂性的模式。牛仔竞技表演也将是关于蛋白质-蛋白质相互作用主题的教育和研究的宝贵工具，以及关于将图形算法应用于生物学中复杂问题的通用资源。竞技场将不仅向对相互作用网络结构感兴趣的高级科学家开放，也将向任何关心更多了解细胞复杂性的生物学学生开放。具体地说，将制定一项培训计划，包括在哈佛医学院免费向国际科学界开放为期两天的“基序发现”培训课程。这项课程将由私人投资公司和高级人员共同讲授这项提议，并提供关于生物学的图形模型和算法的入门课程，以及与竞技表演项目的实践培训。其次，将通过Rodeo门户网站创建一个虚拟培训，其中将包含大量的文档，以及我们的“Motif Discovery”培训研讨会的流媒体视频。第三，将在关于生物计算的两个主要专题讨论会上举办牛仔竞技培训班，该专题讨论会也将通过牛仔竞技表演组织的门户网站提供。最后，竞技表演将被纳入医学院系统生物学系研究生的培训课程。

Harvard University is awarded a grant to build a new publicly available informatics tool, called Rodeo that will allow the integration of data relevant to deciphering protein-protein interaction networks and will provide a number of methods for analyzing such data, including graph-theoretic mathematics. Rodeo will allow the integration and analysis of several databases including a comprehensive set of orthologs from over 50 genomes, a new set of protein-protein (p-p) interaction data from humans together with over 70,000 p-p interactions from other organisms, a massive set of gene expression data from new experiments and experiments hosted by NCBI, and a database of almost 100,000 Expressed Sequence Tags from an organism (the acorn worm) that may hold the key for understanding the developmental complexity which evolved in modern day vertebrate animals. Rodeo will allow the interrogation of data not just from one organism, but across multiple organisms, providing a way to search for convergently evolved patterns a vital step to understanding process. Rodeo will also allow the visualization of motifs to literally 'see' patterns of network complexity. Rodeo will also be a valuable tool for education and research on protein-protein interaction motifs, and a general resource on the application of graph algorithms to complex questions in biology. Rodeo will be made accessible to not just advanced scientists interested in network structures of interactomes, but to any student of biology concerned with learning more about the complexity of the cell. Specifically a training program will be developed consisting of a 2-day training course on 'motif discovery' at Harvard Medical School that will be open to the international scientific community at no cost. This will be co-taught by the PI and senior personnel on the proposal, with introductory lectures on graph models and algorithms for biology, as well as hands-on training with Rodeo. Secondly, a virtual training via the Rodeo web portal will be created that will contain extensive documentation, as well as streaming video of our 'motif discovery' training seminars. Thirdly, Rodeo tutorials will be held at two of the major symposia on biocomputing, which will also be available through the Rodeo web portal. And last not least, Rodeo will be incorporated into the training curriculum for graduate students in the Department of Systems Biology at the Medical School.