Extracting meaningful biological insights from high-throughput mass spectrometry data necessitates minimizing false discoveries. In targeted metabolomics, a key challenge is identifying and filtering false positive metabolic features within the low signal-to-noise ranges of data-independent acquisition (DIA) results. Furthermore, automating assay library creation for DIA analysis and processing extracted ion chromatograms (XICs) has remained problematic. This paper introduces DIAMetAlyzer, a fully automated, open-source workflow combining data-dependent acquisition (DDA) and DIA for library generation, analysis, and statistical validation, rigorously controlling the false discovery rate (FDR) while matching manual analysis in quantification accuracy. An experimentally specific DDA library based on reference substances enables precise compound and marker identification from DIA data, even at low concentrations, facilitating biomarker quantification.

Oliver Alka, Premy Shanthamoorthy, Michael Witting, Karin Kleigrewe, Oliver Kohlbacher, Hannes L. Röst