Optimization of the Identification Rate and Reproducibility of an Untargeted diaPASEF Method Applicable for Quantitative Peptide Profiling of Hydrolyzed Infant Formula

Quality control of hydrolyzed infant formula (HIF) requires comprehensive and precise quantification of its peptide components. Quantitative peptidome analysis by liquid chromatography-tandem mass spectrometry (LC-MS/MS) with data-independent acquisition (DIA) and parallel accumulation-serial fragmentation (PASEF) is used for this application. Here, an optimization strategy was developed to increase the peptide identification rate and the qualitative and quantitative reproducibility of this approach. To expand the peptide identification rate, the originally assigned equidistant ion mobility (IM) windows were transferred to variable ion mobility windows with manually adjusted window placement. To improve the reproducibility, major acquisition parameters, such as the number of diaPASEF scans and ion mobility windows as well as the resulting cycle time, were systematically optimized. Thus, the approach was modified from 17 equidistant windows with a cycle time of 1.8 s to 30 variable windows with a cycle time of 1.7 s. The optimization process led to the identification of 628 peptides versus 522 peptides, increasing the identification rate by 20.3%. Concurrently, the coefficient of variation (CV) for peptide identification was reduced from 10.9 to 0.8%, and for quantitative reproducibility, it was reduced from 24.3 to 17.2%. Based on these results, an optimization workflow is presented to systematically improve the identification rate and reproducibility for other sample types and instruments.