代谢物定性作为代谢组学研究的瓶颈一直以来都是科研热点之一。基于LC-MSn的代谢组学技术在代谢物定性上存在通量低、数据库可参考信息少，代谢物从头定性困难等问题。本项目拟以研究团队多年代谢组学定性工作为基础，构建基于大量代谢物标样和已经定性代谢物的标准化LC-MSn数据库；发展基于数据库中代谢物保留规律和断裂规律的代谢物智能检索定性方法；全面总结数据库信息，构建碎片树、研究保留和化学结构之间关系，结合FT MS，发展基于高分辨精确质量、碎片树和保留时间预测结合的代谢物从头定性方法；应用发展的多层次、全面的代谢物定性策略，构建智能定性系统；并用于胰岛素抵抗代谢特征研究，以期发现胰岛素抵抗相关的新型代谢标志物。拟通过本项目的实施，为小分子代谢物规模化定性提供新策略，提高复杂生物样本代谢物定性的通量和数量，增进对代谢特征的了解，为功能代谢组学发展提供有力的技术支撑。

Metabolites identification is a hotspot issue in metabolomics study. The main bottlenecks include time-consuming, incomplete information in metabolites database, and de novo identification is difficult. The project plans to establish a metabolites database based on identified metabolites of our previous studies and available metabolite standards in the standard operating procedure of LC-MSn. Based on the metabolite database, a metabolite identification method will be developed according metabolites retention time and fragmentation patterns intelligent searching. Furthermore, an integrated identification approach of FT MS, fragmentation trees and quantitative structure–retention relationships (QSRR) was investigated for de novo identification of completely unknown metabolites. The comprehensive metabolite identification strategy will be applied in insulin resistance metabolomics study to discover novel metabolite biomarkers. The major goals of our project are to increase the throughput and to identify more metabolites in complex biological samples, which can enhance the understanding of the metabolic characteristics, and provide strong technical support for the development of functional metabolomics.