【摘要】 针对代谢组学研究中的数据处理问题,本研究建立了基于质谱的数据分析系统MS-IAS(Mass spec-trometry based integrated analysis system)。此系统集成了特征选择、聚类、分类等多种方法,用以处理质谱数据,具有多种统计分析方法能对所选的特征变量进行比较,以发现与所研究问题相关的潜在生物标志物。MS-IAS支持数据与多种算法结果可图形化显示,有助于对数据的解释与分析。以肝病患者的质谱代谢组数据为例,展示MS-IAS的功能,两种特征选择算法从数据集中筛选出了40个对肝病具有区分能力的特征变量,展示了MS-IAS成为代谢组学研究中的通用质谱数据分析系统的潜力。更多还原

【Abstract】 Mass spectrometry(MS) data processing and analysis have become one of the bottlenecks influencing metabolomics applications.In this paper we introduce a software system,MS-IAS(mass spectrometry based integrated analysis system),for processing and analysis of MS-based metabolomics data.It contains various kinds of feature selection,clustering and classification algorithms.Several multiple statistical analysis methods are integrated in the system to compare the selected features,aiming at finding the potential biomarkers.The visualization tools are also very convenient in MS-IAS and helpful to the analysis and interpretation of the data.To demonstrate the usefulness of MS-IAS,one human liver disease metabolomics dataset was used.40 ion features were defined by the two feature selection methods in the system.5 fold cross validation accuracy rates show the validation of the selected features.The results show the potential of MS-IAS to become a general system for MS data processing and analysis in metabolomics.更多还原