WebMeV: A Robust Platform for Intuitive Genomic Data Analysis

项目摘要/摘要： 测序技术的进步使大型群体的基因组和转录组图谱成为可能 研究对象的数量。癌症基因组图谱(TCGA)、DNA元素百科全书等项目 (Encode)、基因类型-组织表达项目(GTEx)等已经将大型、复杂、多个 将基因组数据转移到公共领域。虽然这些大型项目和新测序技术的使用 提供了前所未有的数据量，面临着移动和分析大量数据等技术挑战 多组数据集，以及缺乏直观和易于使用的数据分析工具，限制了广泛的 探索可用的数据，通常将实验生物学家和领域专家与直接 探索数据中的关系。 15年前，我们开始开发MeV，这是一个免费提供的开源软件工具，用于 对基因组数据的直观分析。简单的图形用户界面和丰富的最新状态库 ART分析方法使MeV成为生物信息学中使用最广泛的软件工具之一，有近 自2008年开始统计以来，我们的下载量为260,000次，而 过去几年。尽管MeV取得了成功并继续使用，但我们认识到，大规模、多基因组 使用桌面应用程序不再能轻松分析数据集。为了跟上数据的步伐，我们 需要开发一个利用现代计算技术的新平台，包括基于云的 计算和可扩展的数据存储。 该解决方案由NCI通过ITCR计划(5U01CA151118)提供资金，是一种基于云的Web- 启用的MeV版本(WebMeV；http://mev.tm4.org).WebMeV使用Google Cloud Platform(GCP)及其 利用云计算资源分析大型公共基因组数据的计算引擎基础架构 布景。2016年4月，我们发布了强大的WebMeV版本，并看到该系统的使用不断增长 戏剧性的。该系统已被用于执行超过350,000次分析；目前为WebMeV 每天执行100多次分析，3735名已在系统和该组中注册的用户 每月增长400人(不需要注册)。为了确保广泛使用，我们在网上做了很多 教程，包括为NCI的内部调查人员提供的两个“售罄”教程，WebMeV已成为 基因组分析的关键工具。在这个应用中，我们建议继续维护和改进WebMeV， 通过实现网络推理和表示方法来扩展其功能，以与 癌症基因组学云试点计划，并实施可以促进可再生性的方法 研究。

Project Summary/Abstract: Improvements in sequencing technology have allowed genome and transcriptome profiling of large groups of research subjects. Projects such as The Cancer Genome Atlas (TCGA), the Encyclopedia of DNA Elements (ENCODE), the Genotype-Tissue Expression Project (GTEx), and other have placed large, complex, multi- omic data into the public domain. While these large projects and the use of new sequencing technologies has made an unprecedented quantity of data available, technical challenges such as moving and analyzing large multi-omic data sets, and the lack of intuitive and easy to use tools for data analysis, have limited broad exploration of the available data, often separating experimental biologists and domain experts from directly exploring relationships within the data. More than 15 years ago, we began development of MeV, a freely-available, open source software tool for intuitive analysis of genomic data. The simple graphical user interface and the extensive library of state-of-the- art analytical methods made MeV one of the most widely used software tools in bioinformatics, with nearly 260,000 downloads since we began keeping statistics in 2008 and downloads of nearly 30,000 per year for the past few years. Despite the success of MeV and its continued use, we recognized that large-scale, multi-omic data sets can no longer be analyzed easily using a desktop application. To keep pace with the data, we needed to develop a new platform that draws on modern computing technologies, including cloud-based computing and scalable data storage. The solution, funded by the NCI through the ITCR program (5U01CA151118), is a cloud-based, web- enabled version of MeV (WebMeV; http://mev.tm4.org). WebMeV uses Google Cloud Platform (GCP) and its Compute Engine infrastructure to leverage cloud-computing resources for analyzing large public genomic data sets. In April 2016, we released a robust version of WebMeV and have seen use of the system grow dramatically. The system has already been used to perform more than 350,000 analyses; WebMeV currently performing more than 100 analyses per day, 3,735 users who have registered with the system and that group is growing by 400 per month (registration is not required). To ensure wide use, we have done numerous online tutorials, including two “sold out” tutorials for intramural investigators at the NCI where WebMeV has become a critical tool for genomic analysis. In this application, we propose to continue to maintain and improve WebMeV, to expand its capabilities by implementing methods for network inference and representation, to integrate with the Cancer Genomics Cloud Pilots program, and to implement methods that can advance reproducible research.