Scout-triggered MRM from untargeted data: An R-based strategy for multiplexed targeted metabolomics

The transition from untargeted discovery to targeted validation in metabolomics is a major challenge, often hindered by poor method transferability and retention time (RT) variation between liquid chromatography-mass spectrometry (LC-MS) platforms. To address this, "Scout-MRM Builder," an R-based package for the automated creation of highly multiplexed targeted methods from untargeted high-resolution MS² data, is presented here. A Scout-Triggered Multiple Reaction Monitoring (StMRM) strategy is employed, using N-Alkylpyridinium-3-Sulfonate (NAPS) standards as dynamic RT markers ("scouts"). A specific list of transitions is triggered by the detection of each scout, ensuring robustness against RT shifts. Ion pairs are automatically extracted, scouts are identified, and ready-to-use StMRM methods, including pseudo-MRM transitions for features lacking fragmentation spectra, are generated. From an untargeted analysis of porcine liver extracts, a single StMRM method monitoring 1312 transitions was generated from 558 features. High reproducibility was demonstrated, with 89.9% of detected transitions exhibiting a relative standard deviation (RSD) below 20%. When applied to a model of liver ischemia-reperfusion injury, results highly comparable to the initial untargeted analysis were obtained. A common core of potential biomarkers was identified, with slightly improved statistical performance. In conclusion, the Scout-MRM Builder provides a powerful framework to bridge the gap between discovery and validation, enabling robust targeted analysis at an untargeted scale through enhanced method transferability and reliability.