A Comprehensive Test Suite for SBML

SBML 综合测试套件

• 批准号: 7090186
• 负责人: MICHAEL HUCKA
• 金额: $ 30.27万
• 依托单位: CALIFORNIA INSTITUTE OF TECHNOLOGY
• 依托单位国家: 美国
• 财政年份: 2006
• 资助国家: 美国
• 起止时间: 2006-09-20 至 2008-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/7090186
• 关键词: Internet; bioinformatics; computational biology; computer program /software; computer simulation; data collection methodology /evaluation

DESCRIPTION (provided by applicant): The Systems Biology Markup Language (SBML) is a popular, machine-readable, model representation format allowing computational models of biochemical networks to be exchanged between software tools such as editing programs, simulators, and databases. It is supported not only by open-source and commercial developers of systems biology software, but also by major journals such as Nature. Our group continues to lead and guide the evolution of SBML and related infrastructure, and as a result, we also keenly feel some of the limitations and challenges that have become apparent only as a result of SBML's exploding success. One such challenge is ensuring a high degree of conformance to the SBML specification by the steadily growing number of tools. We have come to realize that simply providing a definition of SBML, holding workshops and discussions, and so forth, are not enough to ensure consistent interpretation and behavior by different software systems. There need to be concrete and objective mechanisms for validating SBML implementations. In preliminary work over the past year, we developed a self-contained "SBML Semantic Test Suite" allowing software developers and users to test the behavior of simulation tools against SBML benchmark models ranging in complexity from basic to advanced, covering a majority of the features defined in the language. In this proposal, we aim to (1) extend and refine this preliminary effort into a comprehensive SBML Test Suite, and (2) develop a web-based system allowing anyone to upload test results, assess them independently, and optionally compare them to the results of other software tools. The overall goal is to provide objective facilities for validating different software tools' coverage and conformance to the definition of SBML, and thereby help ensure the highest level of consistency possible across SBML- aware software systems. This is something that is crucial for users to have confidence in SBML as a standard model representation format for simulation and analysis in biomedical research.

描述（由申请人提供）：系统生物学标记语言（SBML）是一种流行的、机器可读的模型表示格式，允许生物化学网络的计算模型在编辑程序、模拟器和数据库等软件工具之间交换。它不仅得到了系统生物学软件的开源和商业开发者的支持，也得到了《自然》等主要期刊的支持。我们的团队继续领导和指导SBML和相关基础设施的发展，因此，我们也敏锐地感受到一些限制和挑战，这些限制和挑战仅由于SBML的爆炸式成功而变得明显。其中一个挑战是通过不断增长的工具数量来确保与SBML规范的高度一致性。我们已经认识到，简单地提供SBML的定义，举行研讨会和讨论等等，是不足以确保不同软件系统的一致解释和行为的。需要有具体和客观的机制来验证SBML实现。在过去一年的初步工作中，我们开发了一个独立的“SBML语义测试套件”，允许软件开发人员和用户根据复杂程度从基本到高级的SBML基准模型测试模拟工具的行为，涵盖了语言中定义的大部分功能。在这个提案中，我们的目标是(1)扩展和完善这个初步的努力，成为一个全面的SBML测试套件，(2)开发一个基于web的系统，允许任何人上传测试结果，独立评估它们，并有选择地将它们与其他软件工具的结果进行比较。总体目标是为验证不同软件工具的覆盖范围和与SBML定义的一致性提供客观的工具，从而帮助确保在支持SBML的软件系统之间尽可能达到最高级别的一致性。这对于用户信任SBML作为生物医学研究中模拟和分析的标准模型表示格式至关重要。