III: Small: RUI: Investigating Fragmentation Rules and Improving Metabolite Identification Using Graph Grammar and Statistical Methods

III：小：RUI：使用图语法和统计方法研究断裂规则并改进代谢物识别

• 批准号: 2053286
• 负责人: Yingfeng Wang
• 金额: $ 28.41万
• 依托单位: University of Tennessee Chattanooga
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-08-19 至 2025-01-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2053286&HistoricalAwards=false
• 关键词: III; Small; RUI; Investigating; Fragmentation

The proposed work addresses the long-standing challenge of metabolite identification: the lack of fragmentation rules for the large amounts of data from various approaches to analysis. This project can not only advance metabolomics, but also enhance related fields. Tandem mass spectrometry, MS/MS, is a generally used approach, so a range of fields can benefit from improved identification performance, and may be applied to other systems. Moreover, successfully applying graph grammar to solve the graph theory problem in metabolomics will provide support to other scientists who also use graphs to address their research problems by adopting graph grammar methods. A clear focus of this project is to involve undergraduate students in computational research, with the aspects of the overall research project designed at an appropriate level.The specific research objectives of this project is to (1) convert the problem of analyzing MS/MS data into a graph theory problem, (2) use graph grammar and statistical methods to investigate fragmentation rules, and (3) integrate these rules into an identification tool, MIDAS-G. Graph grammar can flexibly represent and process adjacent structures. By taking advantage of its high level for interpretation, graph grammar can work with statistical methods for explicitly investigating fragmentation rules from annotated MS/MS data. These learned rules will in turn be integrated into the identification tool and algorithms will be designed to enable the tool to identify metabolites efficiently.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

拟议的工作解决了代谢物鉴定的长期挑战：来自各种分析方法的大量数据缺乏碎片规则。该项目不仅可以推进代谢组学，还可以增强相关领域。 串联质谱法 (MS/MS) 是一种常用的方法，因此一系列领域都可以从改进的识别性能中受益，并且可以应用于其他系统。此外，成功应用图语法解决代谢组学中的图论问题将为其他也使用图通过采用图语法方法解决其研究问题的科学家提供支持。 该项目的一个明确重点是让本科生参与计算研究，并在适当的水平上设计整个研究项目的各个方面。该项目的具体研究目标是（1）转换分析MS/MS数据的问题转化为图论问题，(2) 使用图语法和统计方法研究碎片规则，(3) 将这些规则集成到识别工具 MIDAS-G 中。图语法可以灵活地表示和处理相邻结构。通过利用其高水平的解释，图语法可以与统计方法一起使用，以明确地研究带注释的 MS/MS 数据的碎片规则。这些学到的规则将被整合到识别工具中，算法的设计将使该工具能够有效地识别代谢物。该奖项反映了 NSF 的法定使命，并通过使用基金会的智力优点和更广泛的影响审查进行评估，被认为值得支持标准。

项目成果

Methods of Metabolite Identification Using MS/MS Data

• DOI: 10.1080/08874417.2019.1681328
• 发表时间: 2019-11
• 期刊: Journal of Computer Information Systems
• 影响因子: 2.8
• 作者: M. Kwak;Kyung-Woo Kang;Yingfeng Wang

A Simple Training Strategy for Graph Autoencoder

• DOI: 10.1145/3383972.3383985
• 发表时间: 2020-02
• 期刊: Proceedings of the 2020 12th International Conference on Machine Learning and Computing
• 作者: Yingfeng Wang;Biyun Xu;M. Kwak;Xiaoqin Zeng

A Machine Learning Based Strategy for Election Result Prediction

• DOI: 10.1109/csci49370.2019.00263
• 发表时间: 2019-12
• 期刊: 2019 International Conference on Computational Science and Computational Intelligence (CSCI)
• 作者: Meng-Hsiu Tsai;Yingfeng Wang;M. Kwak;N. Rigole

Integrating Quantum Computing into De Novo Metabolite Identification

将量子计算集成到从头代谢物识别中

• DOI: 10.54808/jsci.21.02.83
• 发表时间: 2023
• 期刊: Cybernetics and Informatics
• 作者: Tsai, Li-An;Nuckels, Estelle;Wang, Yingfeng
• 通讯作者: Wang, Yingfeng

Metabolite Fragmentation Visualization

代谢物碎片可视化

• DOI: 10.54808/jsci.20.05.138
• 发表时间: 2022
• 期刊: Cybernetics and Informatics
• 作者: Kwak, Myungjae;Molina, Matthew;Arnold, Spencer;Woodward, Andrew;An, Jin-Young;Nuckels, Estelle;Wang, Yingfeng
• 通讯作者: Wang, Yingfeng