Differentiating signals to make biological sense - A guide through databases for MS-based non-targeted metabolomics.
Alberto Gil de la FuenteEmily Grace ArmitageAbraham OteroCoral BarbasJoanna GodzienPublished in: Electrophoresis (2017)
Metabolite identification is one of the most challenging steps in metabolomics studies and reflects one of the greatest bottlenecks in the entire workflow. The success of this step determines the success of the entire research, therefore the quality at which annotations are given requires special attention. A variety of tools and resources are available to aid metabolite identification or annotation, offering different and often complementary functionalities. In preparation for this article, almost 50 databases were reviewed, from which 17 were selected for discussion, chosen for their online ESI-MS functionality. The general characteristics and functions of each database is discussed in turn, considering the advantages and limitations of each along with recommendations for optimal use of each tool, as derived from experiences encountered at the Centre for Metabolomics and Bioanalysis (CEMBIO) in Madrid. These databases were evaluated considering their utility in non-targeted metabolomics, including aspects such as identifier assignment, structural assignment and interpretation of results.
Keyphrases
- mass spectrometry
- ms ms
- liquid chromatography
- multiple sclerosis
- big data
- high resolution
- cancer therapy
- working memory
- emergency department
- social media
- mental health
- machine learning
- magnetic resonance
- clinical practice
- magnetic resonance imaging
- adverse drug
- health information
- rna seq
- artificial intelligence
- sensitive detection
- single cell
- tandem mass spectrometry
- drug delivery