Comparison Of Mass Spectrometry Methods For Untargeted Metabolomics

Untargeted metabolomics is emerging as a powerful and attractive technology,which has been widely used in many research fields,including the discovery of disease biomarkers,the study of metabolic pathways,the monitoring of total environmental exposure,and the search for bioactive chemicals that regulate the phenotype of some diseases.Liquid chromatography high resolution mass spectrometry(LC-HRMS)has become the main technology platform for untargeted metabolomics because of its high sensitivity,good selectivity,wide dynamic range and rich information.In LC-HRMS based untargeted metabolomics,liquid chromatography(LC)is used to separate metabolites according to their chemical properties to reduce the complexity of samples.The retention time information of metabolites can be used to assist the identification of metabolites.After the separation of metabolites,the mass spectrometry was used to detect the abundance of of metabolites signal for quantitative evaluation.At present,there are two common mass spectrometry data acquisition modes in untargeted metabolomics,full scan acquisition(FSA)mode and data dependent acquisition(DDA)mode.The two acquisition methods have their own characteristics and limitations.How to choose the appropriate acquisition mode for a given untargeted metabolomics application is an urgent problem.The purpose of this study is to compare the qualitative and quantitative effects of different modes in untargeted metabolomics,and to provide guidance and basis for how to choose the best data acquisition mode.In this study,we tested the analytical ability of the two data acquisition modes by using a standard mixture of 9 endogenous metabolites and a human urine sample.Metabolomics data acquired by the two data modes were mainly evaluated from five aspects:(1)the number of metabolite features detected,(2)the number of identified metabolites,(3)the intensity distribution of metabolite quantification,(4)the accuracy of metabolite quantification,and(5)the convenience of the method.In the analysis of standard sample,All 9 metabolites could be identified by the two modes.The abundance of 9 metabolites in FSA group was 2.36 times higher than that in DDA group(P<0.05).Among them,the difference of leucine enkephalin(556.2771)was the most obvious,the mean abundance of FSA group was about 3 times higher than that of DDA group,while the difference of valine-tyrosine-valine(380.2185)was smaller,the mean abundance of FSA group was about 2.1 times higher than that of DDA group.The CV of FSA group ranged from 0.81%to 5.42%,and the median CV was 2.48%;The CV of DDA group ranged from 1.87%to 10.09%,and the median CV was 8.75%.For the analysis of urine samples,three different injection volumes were used,and each injection volume was repeated three times.The number of metabolites features detected by FSA mode was more than that detected by DDA mode(P<0.05).The results of metabolite identification showed that the number of metabolites identified by FSA mode was more than that identified by DDA mode.When injecting 1,2 and 4 ?L,the number of metabolite identified by FSA mode was 9.5%,1.9%and 10%higher than those identified by DDA mode,respectively.And the median CV of metabolite abundance analyzed by FSA(7.1%?3.0%and 2.6%)was significantly lower than that by DDA(10.3%,7.0%and 6.5%).When injecting 1,2 and 4 ?L,with the increase of abundance,the median CV of metabolite abundance in FSA mode was 1.3?1.7,1.7?1.8 and 1.4?1.9 times lower than that in DDA mode(P<0.05),and with the increase of peak width,the median CV of FSA mode was 1.5?1.8,1.7?2.1 and 2.0?2.2 times lower than that of DDA mode(P<0.05).In this study,the effect of two different data acquisition modes for mass spectrometry in untargeted metabolomics were compared.Our results show that FSA mode is better than DDA mode in qualitative and quantitative accuracy,however,it requires high system stability,therefore,it is suitable for short-time small-scale sample analysis;DDA mode has lower requirement for system stability,and is convenient and time-saving,which can be used for long-time large-scale sample analysis.Our study provides guidance on how to choose the best data acquisiton mode in the future,in order to give full play to the analysis ability of different modes in untargeted metabolomics.