Optimizing peak picking and development of a suspect screening tool for rapid annotation of the human chemical exposome from LC- HRMS datasets

ABSTRACT: The technological advances of cutting-edge high-resolution mass spectrometry (HRMS) has set the stage for a new paradigm for exposure assessment. However, it is critical to ensure that bioinformatics software developed to process raw HRMS data can disentangle low-abundant xenobiotics from the noise. It is also essential to provide tools to speed up the annotation process. In this study, we optimized and compared the efficiency of open source (e.g. XCMS and MzMine2) and vendor software (e.g. MarkerViewTM and Progenesis QI) to detect low-abundant xenobiotics in human plasma and serum. We show that the rates of false negative can be decreased to below 5% when critical parameters are identified and tuned. Even though similar performances can be achieved using the best tuning for all software, the best detection rate was observed for MzMine2 (ADAP pipeline). We then developed an automatized suspect screening workflow (XenoScreener) based on the mass, three Rt prediction models and isotopic pattern. Indicators were developed to provide intermediate annotation scores for each predictor as well as a global score. With XenoScreener, we show that it is possible to efficiently pre-annotate with a high level of confidence a mix of xenobiotics spiked at real-life concentrations in human plasma and serum samples in a very short time (i.e. less than 3 hours after acquisition). We also demonstrate XenoScreener’s high efficiency for the rapid annotation of various xenobiotics (pharmaceuticals, lifestyle markers, plasticizers, flame-retardant, antifungals) in human plasma and serum using a library of about 2200 xenobiotics (annotation confirmed with MS/MS data).

INSTRUMENT(S): Liquid Chromatography MS - negative - reverse phase, Liquid Chromatography MS - positive - reverse phase

SUBMITTER: Jade Chaker 

PROVIDER: MTBLS1785 | MetaboLights | 2021-01-04

REPOSITORIES: MetaboLights