Dr
Description
The raw data were first converted to mzXML format by MSConvert in the ProteoWizard software package (v3.0.8789) and subsequently processed using XCMS for peak identification, peak filtration, and peak alignment. The XCMS’s default set were as follows: bw=2, ppm=15, peakwidth=c (5, 30), mzwid=0.015, mzdiff=0.01, and method=“centWave”.The metabolites were identified by accuracy mass (molecular weight error<30ppm) and MS/MS data were matched with HMDB (http://www.hmdb.ca), massbank (http://www.massbank.jp/), LipidMaps (http://www.lipidmaps.org), mzclound (https://www.mzcloud.org), Metlin (http://metlin.scripps.edu), and database built by Bionovogene Co., Ltd. (BioNovoGene, Suzhou, Jiangsu, China).After normalization, only ion peaks with relative standard deviations (RSDs) less than 30% in QC samples were kept for metabolite annotation.