Patrick Groos

Chemical derivatization can improve the detectability of analytically challenging compounds, but untargeted feature identification and annotation for chemically derivatized samples are limited by the availability of reference spectra. With the workflow presented in this manuscript, challenges associated with library‑dependent identification of derivatized species are mitigated by in silico derivatization and -MS/MS prediction, which reduce the need for dedicated experimental reference libraries. Furthermore, an improved CCS-value prediction algorithm provides an orthogonal dimension that increases confidence in feature annotation for both native and derivatized compounds. The annotated feature lists can be used for automated generation of target lists for quantitation. Additionally, they can be used for semi‑untargeted stable-isotope labeling analyses, which can be performed, from raw data to results, within the same software environment. Finally, natural isotope abundance correction can be applied on the resulting isotopologue distributions, which allows for a full analysis of stable-isotope labeling data. Consolidation of all processing steps in a single environment minimizes operator intervention, reduces error risk, and makes stable-isotope labeling more widely attainable in metabolomics research. This workflow is exemplified with a nutrient contribution experiment in murine brown adipocytes, presenting substrate integration into metabolic intermediates.