The ability to answer complex biological questions in metabolomics relies on the acquisition of high-quality data. However, due to the complex nature of liquid chromatography-mass spectrometry acquisition, data quality checks are often not done comprehensively and only at the postprocessing step. This can be too late to mitigate analytical problems such as loss of m/z calibration, retention time drift and severe ion suppression. It is often not practically or economically feasible to reanalyze samples, and interpretation of the acquired compromised data, if at all possible, is limited, despite the considerable expenses incurred to obtain them. We therefore introduce QC4Metabolomics, a real-time quality control monitoring software for untargeted metabolomics data. QC4Metabolomics monitors files as they are acquired or retrospectively by tracking any user-defined compound(s) and extracting diagnostic information such as observed m/z, retention time, intensity and peak shape, and presents the results on a web dashboard. QC4Metabolomics also monitors the levels of common or user-defined contaminants. We report herein real-world examples where QC4Metabolomics easily reveals analytical problems retrospectively that could have been immediately addressed with real-time monitoring, so that the samples would have been analyzed without any quality control issues. The Shiny app is available as open-source code at https://github.com/stanstrup/QC4Metabolomics. Docker images and a docker-compose setup file are also provided for easy deployment, along with demo data. The documentation can be found at https://stanstrup.github.io/QC4Metabolomics.
Stanstrup et al. (Tue,) studied this question.