|Title||Data Processing in Metabolomics Capillary Electrophoresis–Mass Spectrometry|
|Publication Type||Book Chapter|
|Year of Publication||2022|
|Authors||Tan, J, Huang, Z-A, Morin, GB, Chen, DDY|
|Book Title||Capillary Electrophoresis‐Mass Spectrometry for Proteomics and Metabolomics|
|Publisher||John Wiley & Sons, Ltd|
|Keywords||CEMS spectra, data extraction, data preprocessing, data processing, metabolite identification, metabolomics, PROTEOMICS, statistical analysis|
Summary Metabolomics data extraction may include peak deconvolution, alignment, and integration. The data extraction from CEMS spectra can usually be completed by a software designed for the extraction of liquid chromatography–mass spectrometry (MS) spectra. The purpose of data preprocessing is to preliminarily adjust the obtained data to facilitate the following statistical analysis. Pre-acquisition normalization is relevant more to experimental setups than data processing, so this chapter discusses post-acquisition normalization, which mainly focuses on the data itself. Statistical analysis is the most important step in the processing of metabolomics data. The chapter also discusses some common statistical methods. One of the most significant differences between the data processing of proteomics and metabolomics is the identification of compounds. Metabolite identification usually starts from searching against databases. The Human Metabolome Database, METLIN database, and MassBank contain comprehensive information for many metabolites, including experimental and predicted MS/MS spectra obtained at multiple collision energies.