MetaProteomics Pipeline (MPP): Integrating a stack of metaproteomics data analysis tools into a full-featured and sustainable bioinformatics workflow

这个项目提案的主要目标是建立一个可持续的、功能齐全的、用户友好的数据分析流程，用于元蛋白质组研究，它集成了现有的三个软件原型。元蛋白质组学是一个不断发展的微生物学领域，通过应用基于质谱学的蛋白质分析，直接从微生物的栖息地研究微生物群落。这一研究领域强烈需要量身定制的软件解决方案，使来自不同背景(微生物学、生态学、医学诊断学等)的研究人员能够参与其中。应用充分和最先进的生物信息学策略来分析高度复杂和不同种类的生物样本。由于这些样本含有数百种潜在未知物种的蛋白质，数据分析和解释属于最具挑战性的任务。这包括对获得的质谱图进行可靠的鉴定，对不明确的(共享多肽的)蛋白质鉴定进行有效的分组，对已鉴定的蛋白质进行准确的量化，以及对分类学和功能元信息进行有意义的整合。为了支持这些任务并克服现有的限制，我们开发了开源软件工具MetaProteomeAnalyzer(MPA)、Pipasic和Prophane。这些原型为元蛋白质组学数据的有意义的分析和解释提供了有效的特征。这个项目的范围是将现有的原型结合到一个单一的端到端解决方案中，为元蛋白质组研究社区提供高可用性、可靠性、互操作性和可发现性。在六个不同的工作包中，我们将重点关注(I)标准工作流定义和实施，(二)使用基准数据的自动化质量保证，(Iii)与其他工具互操作性的数据格式定义，(iv.)无障碍基础设施，(V)文档、培训和改进的可发现性，以及(Vi)传播结果和影响测量。流水线将促进对代谢蛋白质组学数据进行具有时间和成本效益的分析。在元蛋白质组学或生物信息学方面缺乏深入知识的用户将能够进行数据分析，而不会遇到兼容性问题。已经处于早期阶段的领域专家将作为试点用户参与，以评估和验证该管道在元蛋白质组学数据上的表现。此外，将提供一个自动基准，不仅用于评估我们开发的流水线，而且还用于在外地测试其他软件。语义软件注释、文档、培训和在社区平台上分发的方法将使更广泛的研究社区能够访问工作流。来自试点和最终用户的使用统计和调查将帮助我们建立一个可持续的生物信息学渠道，其中考虑到数据分析和工作流程设计的最佳做法。

The main objective of this project proposal is to establish the Metaproteomics Pipeline (MPP), a sustainable, full-featured, and user-friendly data analysis workflow for metaproteomics research which integrates three existing software prototypes.Metaproteomics is an evolving field of microbiology studying microbial communities directly from their habitat by applying mass spectrometry-based protein analytics. This research area strongly demands tailored software solutions that enable researchers from various backgrounds (microbiology, ecology, medical diagnostics etc.) to apply adequate and state-of-the-art bioinformatics strategies for analysing highly complex and heterogeneous biological samples. Since such samples contain proteins of hundreds of potentially unknown species, data analysis and interpretation belong to the most challenging tasks. This includes the reliable identification of acquired mass spectra, the efficient grouping of ambiguous (peptide-sharing) protein identifications, the accurate quantification of identified proteins, and the meaningful integration of taxonomic and functional meta-information. To support these tasks and to overcome existing limitations, we developed the open source software tools MetaProteomeAnalyzer (MPA), Pipasic, and Prophane. These prototypes provide effective features for the meaningful analysis and interpretation of metaproteomics data. The scope of this project is to combine the existing prototypes into a single, end-to-end solution providing high usability, reliability, interoperability, and discoverability to the metaproteomics research community. In six different work packages we will focus on (i.) standard workflow definition and implementation, (ii.) automated quality assurance using benchmarking data, (iii.) data format definition for interoperability with other tools, (iv.) infrastructure for accessibility, (v.) documentation, training, and improved discoverability, and (vi.) dissemination of results and impact measurement.The pipeline will facilitate the time- and cost-efficient analysis of metaproteomics data. Users lacking in-depth knowledge either in metaproteomics or bioinformatics will be able to conduct data analysis without experiencing compatibility issues. Domain experts will be involved as pilot users already at an early stage to evaluate and validate the pipeline with respect to its performance on metaproteomics data. Moreover, an automated benchmarking will be provided not only for evaluating our developed pipeline, but also for testing other software in the field. Methods of semantic software annotation, documentation, training and distribution on community platforms will make the workflow accessible to a wider research community. Usage statistics and surveys from pilot and end users will help us to set up a sustainable bioinformatics pipeline which considers best-practices in data analysis and workflow design.