LEAPS-MPS: Artificial Intelligence Techniques for Automatic NMR Metabolomics Data Processing

该奖项部分由2021年美国救援计划法案（公法117-2）资助。在这个由数学和物理科学理事会和化学部门资助的项目中，北卡罗来纳农业技术州立大学（NCAT）的王博教授和他的学生将开发化学数据分析工具，使非专业人员更容易应用。这些数据与环境和健康研究有关，将通过核磁共振（NMR）方法获得。这种新的数据处理方式将使用人工智能（AI）方法。这项研究通过提供对复杂数据集的快速分析，可能对健康监测产生重大影响。这项研究可能会激励PI所在机构的学生，这是一所历史悠久的黑人学院或大学（HBCU）。王教授计划扩大少数族裔（URM）学生对STEM的参与，包括向学生介绍应用数据科学和人工智能方法，这些都是非常有市场的工作技能，但在本科阶段并不经常教授。将提供独立的跨学科研究项目的机会，并将开发新的课程。使用应用于1D和2D NMR数据分析的AI技术，将改进代谢物的鉴定。核磁共振具有较高的仪器可靠性，但普遍存在数据过拟合的问题。该方法将为复杂光谱的峰提取提供改进的方法。培训数据可从已发表的文献和王教授自己的研究中免费获得，并且该方法将使用王教授小组记录的数据进行测试。所开发的方法将有助于使代谢组学和其他复杂的基于核磁共振的研究对更广泛的科学家开放。本研究将透过课程与课程的开发，以及教育网站的引进，与教育相结合。还将举办夏季讲习班。开普敦大学。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

This award is funded in part under the American Rescue Plan Act of 2021 (Public Law 117-2). In this project, funded by the Mathematical and Physical Sciences Directorate and the Chemistry Division, Professor Bo Wang and his students at North Carolina Agricultural & Technical State University (NCAT) will develop tools for analysis of chemical data that would be easier to apply for non-specialists. The data are related to environmental and health studies and will be obtained using Nuclear Magnetic Resonance (NMR) methods. The novel way of data processing will use artificial intelligence (AI) methods. This research could have significant impacts on health monitoring by providing fast analysis of complicated data sets. The research is likely to inspire students at the PI's institution, which is a Historically Black College or University (HBCU). Prof. Wang’s plans for broadening participation in STEM by underrepresented minority (URM) students include introducing students to applied data science and AI methods, which are very marketable job skills but are not taught regularly at the undergraduate level. Opportunities for independent interdisciplinary research projects will be provided and a new course will be developed.The identification of metabolites will be improved using AI techniques applied to analysis of 1D and 2D NMR data. NMR has high instrument reliability, but there is a common problem of data overfitting. This new method will provide improved ways for extracting peaks in complex spectra. Training data are freely available from published literature and Prof. Wang’s own studies, and the method will be tested with data recorded by Professor Wang’s group. Methods developed will help make metabolomics and other sophisticated NMR-based research accessible to a broader range of scientists. This research will be integrated with education through course and curriculum development, and the introduction of an educational website. Summer workshops will also be conducted. ucted.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.