Five Easy Metrics of Data Quality for LC–MS-Based Global Metabolomics
- 14 September 2020
- journal article
- research article
- Published by American Chemical Society (ACS) in Analytical Chemistry
- Vol. 92 (19) , 12925-12933
Abstract
Data quality in global metabolomics is of great importance for biomarker discovery and systems biology studies. However, comprehensive metrics and methods to evaluate and compare the data quality of global metabolomics data sets are lacking. In this work, we combine newly developed metrics, along with well-known measures, to comprehensively and quantitatively characterize the data quality across two similar LC-MS platforms, with the goal of providing an efficient and improved ability to evaluate the data quality in global metabolite profiling experiments. A pooled human serum sample was run 50 times on two high-resolution LC-QTOF-MS platforms to provide profile and centroid MS data. These data were processed using Progenesis Qi software and then analyzed using five important data quality measures, including retention time drift, number of compounds detected, missing values and MS reproducibility (2 measures). The detected compounds were fit to a gamma distribution versus compound abundance, which was normalized to allow comparison of different platforms. To evaluate missing values, characteristic curves were obtained by plotting the compound detection percentage versus extraction frequency. To characterize reproducibility, the accumulative coefficient of variation (CV) versus percentage of total compounds detected and intra-class correlation coefficient (ICC) versus compound abundance were investigated. Key findings include significantly better performance using profile mode data compared to centroid mode as well quantitatively better performance from the newer, higher resolution instrument. A summary table of results gives a snapshot of the experimental results and provides a template to evaluate the global metabolite profiling workflow. In total, these measures give a good overall view of data quality in global profiling and allow comparisons of data acquisition strategies and platforms as well as optimization of parameters.Keywords
Funding Information
- National Cancer Institute (CA015704)
- University of Washington
- National Institute of Diabetes and Digestive and Kidney Diseases (DK035816, DK119886)
- National Natural Science Foundation of China (81871445)
- National Institute of General Medical Sciences (GM131491, R01GM131491)
- Agilent Technologies, Inc
This publication has 41 references indexed in Scilit:
- Diacetylspermine Is a Novel Prediagnostic Serum Biomarker for Non–Small-Cell Lung Cancer and Has Additive Performance With Pro-Surfactant Protein BJournal of Clinical Oncology, 2015
- The metabolome regulates the epigenetic landscape during naive-to-primed human embryonic stem cell transitionNature Cell Biology, 2015
- Identification of Altered Metabolic Pathways in Plasma and CSF in Mild Cognitive Impairment and Alzheimer’s Disease Using MetabolomicsPLOS ONE, 2013
- An Inhibitor of Mutant IDH1 Delays Growth and Promotes Differentiation of Glioma CellsScience, 2013
- Metabolomics: the apogee of the omics trilogyNature Reviews Molecular Cell Biology, 2012
- Liquid chromatography–mass spectrometry based global metabolite profiling: A reviewAnalytica Chimica Acta, 2012
- Gut flora metabolism of phosphatidylcholine promotes cardiovascular diseaseNature, 2011
- Metabolic oxidation regulates embryonic stem cell differentiationNature Chemical Biology, 2010
- Metabolomic profiles delineate potential role for sarcosine in prostate cancer progressionNature, 2009
- Metabolomics by numbers: acquiring and understanding global metabolite dataTrends in Biotechnology, 2004