MStargetR: a reproducible, containerised workflow for end-to-end targeted (MRM/SRM) mass spectrometry data processing in R

Abstract

Abstract Introduction Targeted metabolic phenotyping by liquid chromatography–tandem mass spectrometry (LC-MS/MS) relies on a fragmented toolchain of proprietary vendor formats, manual integration steps, and ad hoc quality-control (QC) scripts, introducing user- and laboratory-level variation that undermines reproducibility and confounds cross-laboratory and retrospective comparison. Objectives To provide an open-source, R-native workflow for targeted multiple reaction monitoring (MRM/SRM) mass spectrometry data that consolidates vendor file conversion, peak integration, and QC reporting into a single reproducible pipeline while preserving auditable, human-in-the-loop peak review. Methods MStargetR builds on msConvert and Skyline through three modules: msConvertR (vendor-to-mzML conversion), PeakForgeR (peak boundary optimisation and automated peak integration executed through Skyline), and qcCheckR (normalisation, concentration calculation, signal and batch correction, and reporting). Additionally, MStargetR has a standalone correction module and a Shiny graphical user interface. Third-party tools are pinned in version-controlled Docker images (with Apptainer support for high-performance computing), and each analytical plate emits a fully populated sky document for inspection and reimport. Results Applied to a published targeted lipidomics dataset of 128 human plasma samples targeting 1,161 lipid species, MStargetR processed all samples end-to-end, recovering all 1,161 targeted lipid features, 949 of which (81.7%) were detected and returned RSD < 30% across replicated long-term reference QCs. Analysis scaled linearly to 4,200 samples, averaging 4.1 s per sample. Conclusion MStargetR delivers automated batch processing, auditable peak review, and a documented QC layer in a single reproducible pipeline, supporting FAIR-aligned targeted metabolomics.