复杂色谱-质谱联用数据准确快速定性分析新方法研究

Combining the advantages of high-performance separation, high sensitivity and selectivity, the coupled chromatographic and mass spectrometric techniques were extensively employed in various fields such as systems biology, environmental science, food chemistry, medicinal chemistry and so on as the most important approach of instrumental analysis. To date, in contrast to the routine and automated data acquisition steps, the identification requires extensive manual analysis with doubtful results and therefore forms a major bottleneck in data interpretation. Since it could not differentiate the compounds with very similar mass spectra and not sever the identification of compounds out of library, the traditional method of spectral library searches could obviously not satisfy the requirement of automatic identification of unknown compounds in complex samples. In this project, taking human metabolites as research object, we intend to propose a serial of novel methods for automatic identification of unknown compounds after systemic and punctilious investigation to overcome the disadvantages of spectral library searches. Briefly, several parts including proper usage of sample and experiments information, mining mass spectral characteristics, prediction of retention indices and calibration of mass spectra and retention indices will be investigated respectively and combined to build intelligent identification system for routine chromatography mass spectrometric analysis. As a basic science projection, its harvests would benefit the research and applications of chromatography mass spectrometry to a great extent.

色谱质谱联用技术因结合色谱的高效分离能力和质谱的高灵敏度，高选择性等优点广泛应用于系统生物学，环境科学，食品化学，医药化学等等领域。目前，复杂样本中未知化合物的结构鉴定难题已成为制约色谱质谱联用技术广泛应用的瓶颈。传统上数据库搜索的方法已明显不能满足高通量高复杂度样本定性分析的要求。本项目正是针对化合物结构鉴定中存在困难，在原有大量探索工作基础上，以人类代谢物为研究对象，就发展快速准确的数据定性分析方法进行系统而细致的研究。与传统方法中仅把质谱作为指纹谱进行相似度匹配进行结构鉴定相比，本项目拟采用化学计量学方法，从样本与实验信息使用，质谱特征挖掘，色谱保留指数预测，质谱数据和保留指数校正等方面开发新的定性分析方法。这项基础研究的成功完成对于色谱质谱技术的研究和应用将具有十分重要的意义。

色谱质谱联用技术因结合色谱的高效分离能力和质谱的高灵敏度，高选择性等优点广泛应用于系统生物学，环境科学，食品化学，医药化学等等领域。目前，复杂样本中未知化合物的结构鉴定难题已成为制约色谱质谱联用技术广泛应用的瓶颈。传统上数据库搜索的方法已明显不能满足高通量高复杂度样本定性分析的要求。在国家自然科学基金委资助下，本项目建立了生物样品脂肪酸高灵敏全谱分析技术，该技术利用脂肪酸甲酯的质谱裂解规律和保留指数规律，建立了脂肪酸自动化定性定量分析方法，并已应用于代谢组学研究中血浆，尿液中的脂肪酸测定；建立了血浆、尿液代谢物的GC-MS、UPLC-Q/TOF-MS全谱分析方法，人体代谢组色谱质谱智能数据库，氨基酸硅烷化产物质谱、保留时间子数据库。在国际国内学术会议上作报告4次；发表论文20篇，其中SCI论文15篇，单篇最高影响因子达6.47；申请发明专利4项，其中授权专利1项；培育博士研究生1名，硕士研究生5名（含在读研究生2名）。本项目建立的方法和数据库，对于色谱质谱技术的研究和代谢组学中数据处理提供了重要的支撑。