While shotgun lipidomics has advanced the field of lipid analysis, it has limitations including ionization suppression and failure to distinguish isobaric species of possible biological importance. This has led to chromatographic-based lipid profiling approaches using HPLC coupled to high resolution mass spectrometry. For best quantitative results and reproducibility, lipid profiling is done in MS1 mode, however confident lipid annotation requires MS/MS data acquisition to enable product-ion spectral matching against in silico generated databases. While chromatography helps elucidate isomeric lipid species and reduce complexity, it is still not possible to acquire all the MS/MS of interest in a single analysis for complex samples. In this study, a new, fully-automated Q-TOF iterative acquisition mode was used where precursors previously selected for MS/MS fragmentation are excluded on a rolling basis over multiple injections.
Another challenge in lipid profiling is the annotation of lipid MS/MS spectra. Due to the lack of authentic standards and the large diversity of lipid species, MS/MS annotation is done using in silico matching. In this study, a novel software tool was employed which uses a Bayesian-probability-theory algorithm and a theoretical lipid library (LipidBlast) to annotate the iterative mode MS/MS spectra. The tool takes special care not to over-annotate lipid entities by only providing the level of structural information confidently informed by the MS/MS spectra. The tool quickly and automatically generates an accurate mass-retention time database, and the resulting database annotates the MS-only lipid profiling data. This workflow provides more comprehensive lipid annotation than can be achieved by traditional approaches.